From bc8d2a0910b3a49679f6e0c185d894d82e464e2a Mon Sep 17 00:00:00 2001 From: Trevor Cooper Date: Tue, 21 Mar 2017 23:24:54 -0700 Subject: Updated doc locations for new structure Change-Id: I8d948bad350ec90618edac5fc451167c06e8baa5 Signed-off-by: Trevor Cooper --- docs/testing/developer/design/LICENSE | 2 + .../developer/design/factory_and_loader.png | Bin 0 -> 25586 bytes .../developer/design/traffic_controller.png | Bin 0 -> 57868 bytes .../design/trafficgen_integration_guide.rst | 238 +++ docs/testing/developer/design/vsperf.png | Bin 0 -> 93029 bytes .../developer/design/vswitchperf_design.rst | 870 ++++++++++ docs/testing/developer/requirements/LICENSE | 2 + .../developer/requirements/ietf_draft/LICENSE | 12 + .../draft-ietf-bmwg-vswitch-opnfv-00.xml | 1016 ++++++++++++ .../draft-ietf-bmwg-vswitch-opnfv-01.xml | 1027 ++++++++++++ .../draft-vsperf-bmwg-vswitch-opnfv-00.xml | 964 +++++++++++ .../draft-vsperf-bmwg-vswitch-opnfv-01.xml | 1016 ++++++++++++ .../draft-vsperf-bmwg-vswitch-opnfv-02.xml | 1016 ++++++++++++ .../requirements/vm2vm_alternative_benchmark.png | Bin 0 -> 104244 bytes .../developer/requirements/vm2vm_benchmark.png | Bin 0 -> 80797 bytes .../requirements/vm2vm_hypervisor_benchmark.png | Bin 0 -> 122975 bytes .../vm2vm_virtual_interface_benchmark.png | Bin 0 -> 99544 bytes .../developer/requirements/vswitchperf_ltd.rst | 1712 ++++++++++++++++++++ .../developer/requirements/vswitchperf_ltp.rst | 1348 +++++++++++++++ docs/testing/developer/results/results.rst | 38 + docs/testing/developer/results/scenario.rst | 47 + 21 files changed, 9308 insertions(+) create mode 100644 docs/testing/developer/design/LICENSE create mode 100644 docs/testing/developer/design/factory_and_loader.png create mode 100644 docs/testing/developer/design/traffic_controller.png create mode 100644 docs/testing/developer/design/trafficgen_integration_guide.rst create mode 100644 docs/testing/developer/design/vsperf.png create mode 100644 docs/testing/developer/design/vswitchperf_design.rst create mode 100644 docs/testing/developer/requirements/LICENSE create mode 100644 docs/testing/developer/requirements/ietf_draft/LICENSE create mode 100644 docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-00.xml create mode 100644 docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-01.xml create mode 100644 docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-00.xml create mode 100644 docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-01.xml create mode 100644 docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-02.xml create mode 100644 docs/testing/developer/requirements/vm2vm_alternative_benchmark.png create mode 100644 docs/testing/developer/requirements/vm2vm_benchmark.png create mode 100644 docs/testing/developer/requirements/vm2vm_hypervisor_benchmark.png create mode 100644 docs/testing/developer/requirements/vm2vm_virtual_interface_benchmark.png create mode 100644 docs/testing/developer/requirements/vswitchperf_ltd.rst create mode 100644 docs/testing/developer/requirements/vswitchperf_ltp.rst create mode 100644 docs/testing/developer/results/results.rst create mode 100644 docs/testing/developer/results/scenario.rst (limited to 'docs/testing/developer') diff --git a/docs/testing/developer/design/LICENSE b/docs/testing/developer/design/LICENSE new file mode 100644 index 00000000..7bc572ce --- /dev/null +++ b/docs/testing/developer/design/LICENSE @@ -0,0 +1,2 @@ +This work is licensed under a Creative Commons Attribution 4.0 International License. +http://creativecommons.org/licenses/by/4.0 diff --git a/docs/testing/developer/design/factory_and_loader.png b/docs/testing/developer/design/factory_and_loader.png new file mode 100644 index 00000000..290c0af6 Binary files /dev/null and b/docs/testing/developer/design/factory_and_loader.png differ diff --git a/docs/testing/developer/design/traffic_controller.png b/docs/testing/developer/design/traffic_controller.png new file mode 100644 index 00000000..598296ec Binary files /dev/null and b/docs/testing/developer/design/traffic_controller.png differ diff --git a/docs/testing/developer/design/trafficgen_integration_guide.rst b/docs/testing/developer/design/trafficgen_integration_guide.rst new file mode 100644 index 00000000..382cedcb --- /dev/null +++ b/docs/testing/developer/design/trafficgen_integration_guide.rst @@ -0,0 +1,238 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intel Corporation, AT&T and others. + +=================================== +Traffic Generator Integration Guide +=================================== + +Intended Audience +================= + +This document is intended to aid those who want to integrate new traffic +generator into the vsperf code. It is expected, that reader has already +read generic part of :ref:`vsperf-design`. + +Let us create a sample traffic generator called **sample_tg**, step by step. + +Step 1 - create a directory +=========================== + +Implementation of trafficgens is located at tools/pkt_gen/ directory, +where every implementation has its dedicated sub-directory. It is +required to create a new directory for new traffic generator +implementations. + +E.g. + +.. code-block:: console + + $ mkdir tools/pkt_gen/sample_tg + +Step 2 - create a trafficgen module +=================================== + +Every trafficgen class must inherit from generic **ITrafficGenerator** +interface class. VSPERF during its initialization scans content of pkt_gen +directory for all python modules, that inherit from **ITrafficGenerator**. These +modules are automatically added into the list of supported traffic generators. + +Example: + +Let us create a draft of tools/pkt_gen/sample_tg/sample_tg.py module. + +.. code-block:: python + + from tools.pkt_gen import trafficgen + + class SampleTG(trafficgen.ITrafficGenerator): + """ + A sample traffic generator implementation + """ + pass + +VSPERF is immediately aware of the new class: + +.. code-block:: console + + $ ./vsperf --list-trafficgen + +Output should look like: + +.. code-block:: console + + Classes derived from: ITrafficGenerator + ====== + + * Ixia: A wrapper around the IXIA traffic generator. + + * IxNet: A wrapper around IXIA IxNetwork applications. + + * Dummy: A dummy traffic generator whose data is generated by the user. + + * SampleTG: A sample traffic generator implementation + + * TestCenter: Spirent TestCenter + + +Step 3 - configuration +====================== + +All configuration values, required for correct traffic generator function, are passed +from VSPERF to the traffic generator in a dictionary. Default values shared among +all traffic generators are defined in **conf/03_traffic.conf** within **TRAFFIC** +dictionary. Default values are loaded by **ITrafficGenerator** interface class +automatically, so it is not needed to load them explicitly. In case that there are +any traffic generator specific default values, then they should be set within class +specific **__init__** function. + +VSPERF passes test specific configuration within **traffic** dictionary to every +start and send function. So implementation of these functions must ensure, +that default values are updated with the testcase specific values. Proper merge +of values is assured by call of **merge_spec** function from **conf** module. + +Example of **merge_spec** usage in **tools/pkt_gen/sample_tg/sample_tg.py** module: + +.. code-block:: python + + from conf import merge_spec + + def start_rfc2544_throughput(self, traffic=None, duration=30): + self._params = {} + self._params['traffic'] = self.traffic_defaults.copy() + if traffic: + self._params['traffic'] = merge_spec( + self._params['traffic'], traffic) + + +Step 4 - generic functions +========================== + +There are some generic functions, which every traffic generator should provide. +Although these functions are mainly optional, at least empty implementation must +be provided. This is required, so that developer is explicitly aware of these +functions. + +The **connect** function is called from the traffic generator controller from its +**__enter__** method. This function should assure proper connection initialization +between DUT and traffic generator. In case, that such implementation is not needed, +empty implementation is required. + +The **disconnect** function should perform clean up of any connection specific +actions called from the **connect** function. + +Example in **tools/pkt_gen/sample_tg/sample_tg.py** module: + +.. code-block:: python + + def connect(self): + pass + + def disconnect(self): + pass + +.. _step-5-supported-traffic-types: + +Step 5 - supported traffic types +================================ + +Currently VSPERF supports three different types of tests for traffic generators, +these are identified in vsperf through the traffic type, which include: + + * RFC2544 throughput - Send fixed size packets at different rates, using + traffic configuration, until minimum rate at which no packet loss is + detected is found. Methods with its implementation have suffix + **_rfc2544_throughput**. + + * RFC2544 back2back - Send fixed size packets at a fixed rate, using traffic + configuration, for specified time interval. Methods with its + implementation have suffix **_rfc2544_back2back**. + + * continuous flow - Send fixed size packets at given framerate, using traffic + configuration, for specified time interval. Methods with its + implementation have suffix **_cont_traffic**. + +In general, both synchronous and asynchronous interfaces must be implemented +for each traffic type. Synchronous functions start with prefix **send_**. +Asynchronous with prefixes **start_** and **wait_** in case of throughput +and back2back and **start_** and **stop_** in case of continuous traffic type. + +Example of synchronous interfaces: + +.. code-block:: python + + def send_rfc2544_throughput(self, traffic=None, tests=1, duration=20, + lossrate=0.0): + def send_rfc2544_back2back(self, traffic=None, tests=1, duration=20, + lossrate=0.0): + def send_cont_traffic(self, traffic=None, duration=20): + +Example of asynchronous interfaces: + +.. code-block:: python + + def start_rfc2544_throughput(self, traffic=None, tests=1, duration=20, + lossrate=0.0): + def wait_rfc2544_throughput(self): + + def start_rfc2544_back2back(self, traffic=None, tests=1, duration=20, + lossrate=0.0): + def wait_rfc2544_back2back(self): + + def start_cont_traffic(self, traffic=None, duration=20): + def stop_cont_traffic(self): + +Description of parameters used by **send**, **start**, **wait** and **stop** +functions: + + * param **traffic**: A dictionary with detailed definition of traffic + pattern. It contains following parameters to be implemented by + traffic generator. + + Note: Traffic dictionary has also virtual switch related parameters, + which are not listed below. + + Note: There are parameters specific to testing of tunnelling protocols, + which are discussed in detail at :ref:`integration-tests` userguide. + + * param **traffic_type**: One of the supported traffic types, + e.g. **rfc2544_throughput**, **rfc2544_continuous** + or **rfc2544_back2back**. + * param **frame_rate**: Defines desired percentage of frame + rate used during continuous stream tests. + * param **bidir**: Specifies if generated traffic will be full-duplex + (true) or half-duplex (false). + * param **multistream**: Defines number of flows simulated by traffic + generator. Value 0 disables MultiStream feature. + * param **stream_type**: Stream Type defines ISO OSI network layer + used for simulation of multiple streams. + Supported values: + + * **L2** - iteration of destination MAC address + * **L3** - iteration of destination IP address + * **L4** - iteration of destination port of selected transport protocol + + * param **l2**: A dictionary with data link layer details, e.g. **srcmac**, + **dstmac** and **framesize**. + * param **l3**: A dictionary with network layer details, e.g. **srcip**, + **dstip** and **proto**. + * param **l3**: A dictionary with transport layer details, e.g. **srcport**, + **dstport**. + * param **vlan**: A dictionary with vlan specific parameters, + e.g. **priority**, **cfi**, **id** and vlan on/off switch **enabled**. + + * param **tests**: Number of times the test is executed. + * param **duration**: Duration of continuous test or per iteration duration + in case of RFC2544 throughput or back2back traffic types. + * param **lossrate**: Acceptable lossrate percentage. + +Step 6 - passing back results +============================= + +It is expected that methods **send**, **wait** and **stop** will return +values measured by traffic generator within a dictionary. Dictionary keys +are defined in **ResultsConstants** implemented in +**core/results/results_constants.py**. Please check sections for RFC2544 +Throughput & Continuous and for Back2Back. The same key names should +be used by all traffic generator implementations. + diff --git a/docs/testing/developer/design/vsperf.png b/docs/testing/developer/design/vsperf.png new file mode 100644 index 00000000..4af2ac62 Binary files /dev/null and b/docs/testing/developer/design/vsperf.png differ diff --git a/docs/testing/developer/design/vswitchperf_design.rst b/docs/testing/developer/design/vswitchperf_design.rst new file mode 100644 index 00000000..da7ec6fd --- /dev/null +++ b/docs/testing/developer/design/vswitchperf_design.rst @@ -0,0 +1,870 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intel Corporation, AT&T and others. + +.. _vsperf-design: + +====================== +VSPERF Design Document +====================== + +Intended Audience +================= + +This document is intended to aid those who want to modify the vsperf code. Or +to extend it - for example to add support for new traffic generators, +deployment scenarios and so on. + +Usage +===== + +Example Connectivity to DUT +--------------------------- + +Establish connectivity to the VSPERF DUT Linux host, such as the DUT in Pod 3, +by following the steps in `Testbed POD3 +`__ + +The steps cover booking the DUT and establishing the VSPERF environment. + +Example Command Lines +--------------------- + +List all the cli options: + +.. code-block:: console + + $ ./vsperf -h + +Run all tests that have ``tput`` in their name - ``phy2phy_tput``, ``pvp_tput`` etc.: + +.. code-block:: console + + $ ./vsperf --tests 'tput' + +As above but override default configuration with settings in '10_custom.conf'. +This is useful as modifying configuration directly in the configuration files +in ``conf/NN_*.py`` shows up as changes under git source control: + +.. code-block:: console + + $ ./vsperf --conf-file=/10_custom.conf --tests 'tput' + +Override specific test parameters. Useful for shortening the duration of tests +for development purposes: + +.. code-block:: console + + $ ./vsperf --test-params 'TRAFFICGEN_DURATION=10;TRAFFICGEN_RFC2544_TESTS=1;' \ + 'TRAFFICGEN_PKT_SIZES=(64,)' pvp_tput + +Typical Test Sequence +===================== + +This is a typical flow of control for a test. + +.. image:: vsperf.png + +.. _design-configuration: + +Configuration +============= + +The conf package contains the configuration files (``*.conf``) for all system +components, it also provides a ``settings`` object that exposes all of these +settings. + +Settings are not passed from component to component. Rather they are available +globally to all components once they import the conf package. + +.. code-block:: python + + from conf import settings + ... + log_file = settings.getValue('LOG_FILE_DEFAULT') + +Settings files (``*.conf``) are valid python code so can be set to complex +types such as lists and dictionaries as well as scalar types: + +.. code-block:: python + + first_packet_size = settings.getValue('PACKET_SIZE_LIST')[0] + +Configuration Procedure and Precedence +-------------------------------------- + +Configuration files follow a strict naming convention that allows them to be +processed in a specific order. All the .conf files are named ``NN_name.conf``, +where NN is a decimal number. The files are processed in order from 00_name.conf +to 99_name.conf so that if the name setting is given in both a lower and higher +numbered conf file then the higher numbered file is the effective setting as it +is processed after the setting in the lower numbered file. + +The values in the file specified by ``--conf-file`` takes precedence over all +the other configuration files and does not have to follow the naming +convention. + +.. _paths-documentation: + +Configuration of PATHS dictionary +--------------------------------- + +VSPERF uses external tools like Open vSwitch and Qemu for execution of testcases. These +tools may be downloaded and built automatically (see :ref:`vsperf-installation-script`) +or installed manually by user from binary packages. It is also possible to use a combination +of both approaches, but it is essential to correctly set paths to all required tools. +These paths are stored within a PATHS dictionary, which is evaluated before execution +of each testcase, in order to setup testcase specific environment. Values selected for testcase +execution are internally stored inside TOOLS dictionary, which is used by VSPERF to execute +external tools, load kernel modules, etc. + +The default configuration of PATHS dictionary is spread among three different configuration files +to follow logical grouping of configuration options. Basic description of PATHS dictionary +is placed inside ``conf/00_common.conf``. The configuration specific to DPDK and vswitches +is located at ``conf/02_vswitch.conf``. The last part related to the Qemu is defined inside +``conf/04_vnf.conf``. Default configuration values can be used in case, that all required +tools were downloaded and built automatically by vsperf itself. In case, that some of +tools were installed manually from binary packages, then it will be necessary to modify +the content of PATHS dictionary accordingly. + +Dictionary has a specific section of configuration options for every tool type, it means: + + * ``PATHS['vswitch']`` - contains a separete dictionary for each of vswitches supported by VSPEF + + Example: + + .. code-block:: python + + PATHS['vswitch'] = { + 'OvsDpdkVhost': { ... }, + 'OvsVanilla' : { ... }, + ... + } + + * ``PATHS['dpdk']`` - contains paths to the dpdk sources, kernel modules and tools (e.g. testpmd) + + Example: + + .. code-block:: python + + PATHS['dpdk'] = { + 'type' : 'src', + 'src': { + 'path': os.path.join(ROOT_DIR, 'src/dpdk/dpdk/'), + 'modules' : ['uio', os.path.join(RTE_TARGET, 'kmod/igb_uio.ko')], + 'bind-tool': 'tools/dpdk*bind.py', + 'testpmd': os.path.join(RTE_TARGET, 'app', 'testpmd'), + }, + ... + } + + * ``PATHS['qemu']`` - contains paths to the qemu sources and executable file + + Example: + + .. code-block:: python + + PATHS['qemu'] = { + 'type' : 'bin', + 'bin': { + 'qemu-system': 'qemu-system-x86_64' + }, + ... + } + +Every section specific to the particular vswitch, dpdk or qemu may contain following types +of configuration options: + + * option ``type`` - is a string, which defines the type of configured paths ('src' or 'bin') + to be selected for a given section: + + * value ``src`` means, that VSPERF will use vswitch, DPDK or QEMU built from sources + e.g. by execution of ``systems/build_base_machine.sh`` script during VSPERF + installation + + * value ``bin`` means, that VSPERF will use vswitch, DPDK or QEMU binaries installed + directly in the operating system, e.g. via OS specific packaging system + + * option ``path`` - is a string with a valid system path; Its content is checked for + existence, prefixed with section name and stored into TOOLS for later use + e.g. ``TOOLS['dpdk_src']`` or ``TOOLS['vswitch_src']`` + + * option ``modules`` - is list of strings with names of kernel modules; Every module name + from given list is checked for a '.ko' suffix. In case that it matches and if it is not + an absolute path to the module, then module name is prefixed with value of ``path`` + option defined for the same section + + Example: + + .. code-block:: python + + """ + snippet of PATHS definition from the configuration file: + """ + PATHS['vswitch'] = { + 'OvsVanilla' = { + 'type' : 'src', + 'src': { + 'path': '/tmp/vsperf/src_vanilla/ovs/ovs/', + 'modules' : ['datapath/linux/openvswitch.ko'], + ... + }, + ... + } + ... + } + + """ + Final content of TOOLS dictionary used during runtime: + """ + TOOLS['vswitch_modules'] = ['/tmp/vsperf/src_vanilla/ovs/ovs/datapath/linux/openvswitch.ko'] + + * all other options are strings with names and paths to specific tools; If a given string + contains a relative path and option ``path`` is defined for a given section, then string + content will be prefixed with content of the ``path``. Otherwise the name of the tool will be + searched within standard system directories. In case that filename contains OS specific + wildcards, then they will be expanded to the real path. At the end of the processing, every + absolute path will be checked for its existence. In case that temporary path (i.e. path with + a ``_tmp`` suffix) does not exist, then log will be written and vsperf will continue. If any + other path will not exist, then vsperf execution will be terminated with a runtime error. + + Example: + + .. code-block:: python + + """ + snippet of PATHS definition from the configuration file: + """ + PATHS['vswitch'] = { + 'OvsDpdkVhost': { + 'type' : 'src', + 'src': { + 'path': '/tmp/vsperf/src_vanilla/ovs/ovs/', + 'ovs-vswitchd': 'vswitchd/ovs-vswitchd', + 'ovsdb-server': 'ovsdb/ovsdb-server', + ... + } + ... + } + ... + } + + """ + Final content of TOOLS dictionary used during runtime: + """ + TOOLS['ovs-vswitchd'] = '/tmp/vsperf/src_vanilla/ovs/ovs/vswitchd/ovs-vswitchd' + TOOLS['ovsdb-server'] = '/tmp/vsperf/src_vanilla/ovs/ovs/ovsdb/ovsdb-server' + +Note: In case that ``bin`` type is set for DPDK, then ``TOOLS['dpdk_src']`` will be set to +the value of ``PATHS['dpdk']['src']['path']``. The reason is, that VSPERF uses downloaded +DPDK sources to copy DPDK and testpmd into the GUEST, where testpmd is built. In case, +that DPDK sources are not available, then vsperf will continue with test execution, +but testpmd can't be used as a guest loopback. This is useful in case, that other guest +loopback applications (e.g. buildin or l2fwd) are used. + +Note: In case of RHEL 7.3 OS usage, binary package configuration is required +for Vanilla OVS tests. With the installation of a supported rpm for OVS there is +a section in the ``conf\10_custom.conf`` file that can be used. + +.. _configuration-of-traffic-dictionary: + +Configuration of TRAFFIC dictionary +----------------------------------- + +TRAFFIC dictionary is used for configuration of traffic generator. Default values +can be found in configuration file ``conf/03_traffic.conf``. These default values +can be modified by (first option has the highest priorty): + + 1. ``Parameters`` section of testcase defintion + 2. command line options specified by ``--test-params`` argument + 3. custom configuration file + +It is to note, that in case of option 1 and 2, it is possible to specify only +values, which should be changed. In case of custom configuration file, it is +required to specify whole ``TRAFFIC`` dictionary with its all values or explicitly +call and update() method of ``TRAFFIC`` dictionary. + +Detailed description of ``TRAFFIC`` dictionary items follows: + +.. code-block:: console + + 'traffic_type' - One of the supported traffic types. + E.g. rfc2544_throughput, rfc2544_back2back + or rfc2544_continuous + Data type: str + Default value: "rfc2544_throughput". + 'bidir' - Specifies if generated traffic will be full-duplex (True) + or half-duplex (False) + Data type: str + Supported values: "True", "False" + Default value: "False". + 'frame_rate' - Defines desired percentage of frame rate used during + continuous stream tests. + Data type: int + Default value: 100. + 'multistream' - Defines number of flows simulated by traffic generator. + Value 0 disables multistream feature + Data type: int + Supported values: 0-65535 + Default value: 0. + 'stream_type' - Stream type is an extension of the "multistream" feature. + If multistream is disabled, then stream type will be + ignored. Stream type defines ISO OSI network layer used + for simulation of multiple streams. + Data type: str + Supported values: + "L2" - iteration of destination MAC address + "L3" - iteration of destination IP address + "L4" - iteration of destination port + of selected transport protocol + Default value: "L4". + 'pre_installed_flows' + - Pre-installed flows is an extension of the "multistream" + feature. If enabled, it will implicitly insert a flow + for each stream. If multistream is disabled, then + pre-installed flows will be ignored. + Note: It is supported only for p2p deployment scenario. + Data type: str + Supported values: + "Yes" - flows will be inserted into OVS + "No" - flows won't be inserted into OVS + Default value: "No". + 'flow_type' - Defines flows complexity. + Data type: str + Supported values: + "port" - flow is defined by ingress ports + "IP" - flow is defined by ingress ports + and src and dst IP addresses + Default value: "port" + 'l2' - A dictionary with l2 network layer details. Supported + values are: + 'srcmac' - Specifies source MAC address filled by traffic generator. + NOTE: It can be modified by vsperf in some scenarios. + Data type: str + Default value: "00:00:00:00:00:00". + 'dstmac' - Specifies destination MAC address filled by traffic generator. + NOTE: It can be modified by vsperf in some scenarios. + Data type: str + Default value: "00:00:00:00:00:00". + 'framesize' - Specifies default frame size. This value should not be + changed directly. It will be overridden during testcase + execution by values specified by list TRAFFICGEN_PKT_SIZES. + Data type: int + Default value: 64 + 'l3' - A dictionary with l3 network layer details. Supported + values are: + 'srcip' - Specifies source MAC address filled by traffic generator. + NOTE: It can be modified by vsperf in some scenarios. + Data type: str + Default value: "1.1.1.1". + 'dstip' - Specifies destination MAC address filled by traffic generator. + NOTE: It can be modified by vsperf in some scenarios. + Data type: str + Default value: "90.90.90.90". + 'proto' - Specifies deflaut protocol type. + Please check particular traffic generator implementation + for supported protocol types. + Data type: str + Default value: "udp". + 'l4' - A dictionary with l4 network layer details. Supported + values are: + 'srcport' - Specifies source port of selected transport protocol. + NOTE: It can be modified by vsperf in some scenarios. + Data type: int + Default value: 3000 + 'dstport' - Specifies destination port of selected transport protocol. + NOTE: It can be modified by vsperf in some scenarios. + Data type: int + Default value: 3001 + 'vlan' - A dictionary with vlan encapsulation details. Supported + values are: + 'enabled' - Specifies if vlan encapsulation should be enabled or + disabled. + Data type: bool + Default value: False + 'id' - Specifies vlan id. + Data type: int (NOTE: must fit to 12 bits) + Default value: 0 + 'priority' - Specifies a vlan priority (PCP header field). + Data type: int (NOTE: must fit to 3 bits) + Default value: 0 + 'cfi' - Specifies if frames can or cannot be dropped during + congestion (DEI header field). + Data type: int (NOTE: must fit to 1 bit) + Default value: 0 + +.. _configuration-of-guest-options: + +Configuration of GUEST options +------------------------------ + +VSPERF is able to setup scenarios involving a number of VMs in series or in parallel. +All configuration options related to a particular VM instance are defined as +lists and prefixed with ``GUEST_`` label. It is essential, that there is enough +items in all ``GUEST_`` options to cover all VM instances involved in the test. +In case there is not enough items, then VSPERF will use the first item of +particular ``GUEST_`` option to expand the list to required length. + +Example of option expansion for 4 VMs: + + .. code-block:: python + + """ + Original values: + """ + GUEST_SMP = ['2'] + GUEST_MEMORY = ['2048', '4096'] + + """ + Values after automatic expansion: + """ + GUEST_SMP = ['2', '2', '2', '2'] + GUEST_MEMORY = ['2048', '4096', '2048', '2048'] + + +First option can contain macros starting with ``#`` to generate VM specific values. +These macros can be used only for options of ``list`` or ``str`` types with ``GUEST_`` +prefix. + +Example of macros and their expnasion for 2 VMs: + + .. code-block:: python + + """ + Original values: + """ + GUEST_SHARE_DIR = ['/tmp/qemu#VMINDEX_share'] + GUEST_BRIDGE_IP = ['#IP(1.1.1.5)/16'] + + """ + Values after automatic expansion: + """ + GUEST_SHARE_DIR = ['/tmp/qemu0_share', '/tmp/qemu1_share'] + GUEST_BRIDGE_IP = ['1.1.1.5/16', '1.1.1.6/16'] + +Additional examples are available at ``04_vnf.conf``. + +Note: In case, that macro is detected in the first item of the list, then +all other items are ignored and list content is created automatically. + +Multiple macros can be used inside one configuration option definition, but macros +cannot be used inside other macros. The only exception is macro ``#VMINDEX``, which +is expanded first and thus it can be used inside other macros. + +Following macros are supported: + + * ``#VMINDEX`` - it is replaced by index of VM being executed; This macro + is expanded first, so it can be used inside other macros. + + Example: + + .. code-block:: python + + GUEST_SHARE_DIR = ['/tmp/qemu#VMINDEX_share'] + + * ``#MAC(mac_address[, step])`` - it will iterate given ``mac_address`` + with optional ``step``. In case that step is not defined, then it is set to 1. + It means, that first VM will use the value of ``mac_address``, second VM + value of ``mac_address`` increased by ``step``, etc. + + Example: + + .. code-block:: python + + GUEST_NICS = [[{'mac' : '#MAC(00:00:00:00:00:01,2)'}]] + + * ``#IP(ip_address[, step])`` - it will iterate given ``ip_address`` + with optional ``step``. In case that step is not defined, then it is set to 1. + It means, that first VM will use the value of ``ip_address``, second VM + value of ``ip_address`` increased by ``step``, etc. + + Example: + + .. code-block:: python + + GUEST_BRIDGE_IP = ['#IP(1.1.1.5)/16'] + + * ``#EVAL(expression)`` - it will evaluate given ``expression`` as python code; + Only simple expressions should be used. Call of the functions is not supported. + + Example: + + .. code-block:: python + + GUEST_CORE_BINDING = [('#EVAL(6+2*#VMINDEX)', '#EVAL(7+2*#VMINDEX)')] + +Other Configuration +------------------- + +``conf.settings`` also loads configuration from the command line and from the environment. + +.. _pxp-deployment: + +PXP Deployment +============== + +Every testcase uses one of the supported deployment scenarios to setup test environment. +The controller responsible for a given scenario configures flows in the vswitch to route +traffic among physical interfaces connected to the traffic generator and virtual +machines. VSPERF supports several deployments including PXP deployment, which can +setup various scenarios with multiple VMs. + +These scenarios are realized by VswitchControllerPXP class, which can configure and +execute given number of VMs in serial or parallel configurations. Every VM can be +configured with just one or an even number of interfaces. In case that VM has more than +2 interfaces, then traffic is properly routed among pairs of interfaces. + +Example of traffic routing for VM with 4 NICs in serial configuration: + +.. code-block:: console + + +------------------------------------------+ + | VM with 4 NICs | + | +---------------+ +---------------+ | + | | Application | | Application | | + | +---------------+ +---------------+ | + | ^ | ^ | | + | | v | v | + | +---------------+ +---------------+ | + | | logical ports | | logical ports | | + | | 0 1 | | 2 3 | | + +--+---------------+----+---------------+--+ + ^ : ^ : + | | | | + : v : v + +-----------+---------------+----+---------------+----------+ + | vSwitch | 0 1 | | 2 3 | | + | | logical ports | | logical ports | | + | previous +---------------+ +---------------+ next | + | VM or PHY ^ | ^ | VM or PHY| + | port -----+ +------------+ +---> port | + +-----------------------------------------------------------+ + +It is also possible to define different number of interfaces for each VM to better +simulate real scenarios. + +Example of traffic routing for 2 VMs in serial configuration, where 1st VM has +4 NICs and 2nd VM 2 NICs: + +.. code-block:: console + + +------------------------------------------+ +---------------------+ + | 1st VM with 4 NICs | | 2nd VM with 2 NICs | + | +---------------+ +---------------+ | | +---------------+ | + | | Application | | Application | | | | Application | | + | +---------------+ +---------------+ | | +---------------+ | + | ^ | ^ | | | ^ | | + | | v | v | | | v | + | +---------------+ +---------------+ | | +---------------+ | + | | logical ports | | logical ports | | | | logical ports | | + | | 0 1 | | 2 3 | | | | 0 1 | | + +--+---------------+----+---------------+--+ +--+---------------+--+ + ^ : ^ : ^ : + | | | | | | + : v : v : v + +-----------+---------------+----+---------------+-------+---------------+----------+ + | vSwitch | 0 1 | | 2 3 | | 4 5 | | + | | logical ports | | logical ports | | logical ports | | + | previous +---------------+ +---------------+ +---------------+ next | + | VM or PHY ^ | ^ | ^ | VM or PHY| + | port -----+ +------------+ +---------------+ +----> port | + +-----------------------------------------------------------------------------------+ + +The number of VMs involved in the test and the type of their connection is defined +by deployment name as follows: + + * ``pvvp[number]`` - configures scenario with VMs connected in series with + optional ``number`` of VMs. In case that ``number`` is not specified, then + 2 VMs will be used. + + Example of 2 VMs in a serial configuration: + + .. code-block:: console + + +----------------------+ +----------------------+ + | 1st VM | | 2nd VM | + | +---------------+ | | +---------------+ | + | | Application | | | | Application | | + | +---------------+ | | +---------------+ | + | ^ | | | ^ | | + | | v | | | v | + | +---------------+ | | +---------------+ | + | | logical ports | | | | logical ports | | + | | 0 1 | | | | 0 1 | | + +---+---------------+--+ +---+---------------+--+ + ^ : ^ : + | | | | + : v : v + +---+---------------+---------+---------------+--+ + | | 0 1 | | 3 4 | | + | | logical ports | vSwitch | logical ports | | + | +---------------+ +---------------+ | + | ^ | ^ | | + | | +-----------------+ v | + | +----------------------------------------+ | + | | physical ports | | + | | 0 1 | | + +---+----------------------------------------+---+ + ^ : + | | + : v + +------------------------------------------------+ + | | + | traffic generator | + | | + +------------------------------------------------+ + + * ``pvpv[number]`` - configures scenario with VMs connected in parallel with + optional ``number`` of VMs. In case that ``number`` is not specified, then + 2 VMs will be used. Multistream feature is used to route traffic to particular + VMs (or NIC pairs of every VM). It means, that VSPERF will enable multistream + feaure and sets the number of streams to the number of VMs and their NIC + pairs. Traffic will be dispatched based on Stream Type, i.e. by UDP port, + IP address or MAC address. + + Example of 2 VMs in a parallel configuration, where traffic is dispatched + based on the UDP port. + + .. code-block:: console + + +----------------------+ +----------------------+ + | 1st VM | | 2nd VM | + | +---------------+ | | +---------------+ | + | | Application | | | | Application | | + | +---------------+ | | +---------------+ | + | ^ | | | ^ | | + | | v | | | v | + | +---------------+ | | +---------------+ | + | | logical ports | | | | logical ports | | + | | 0 1 | | | | 0 1 | | + +---+---------------+--+ +---+---------------+--+ + ^ : ^ : + | | | | + : v : v + +---+---------------+---------+---------------+--+ + | | 0 1 | | 3 4 | | + | | logical ports | vSwitch | logical ports | | + | +---------------+ +---------------+ | + | ^ | ^ : | + | | ......................: : | + | UDP | UDP : | : | + | port| port: +--------------------+ : | + | 0 | 1 : | : | + | | : v v | + | +----------------------------------------+ | + | | physical ports | | + | | 0 1 | | + +---+----------------------------------------+---+ + ^ : + | | + : v + +------------------------------------------------+ + | | + | traffic generator | + | | + +------------------------------------------------+ + + +PXP deployment is backward compatible with PVP deployment, where ``pvp`` is +an alias for ``pvvp1`` and it executes just one VM. + +The number of interfaces used by VMs is defined by configuration option +``GUEST_NICS_NR``. In case that more than one pair of interfaces is defined +for VM, then: + + * for ``pvvp`` (serial) scenario every NIC pair is connected in serial + before connection to next VM is created + * for ``pvpv`` (parallel) scenario every NIC pair is directly connected + to the physical ports and unique traffic stream is assigned to it + +Examples: + + * Deployment ``pvvp10`` will start 10 VMs and connects them in series + * Deployment ``pvpv4`` will start 4 VMs and connects them in parallel + * Deployment ``pvpv1`` and GUEST_NICS_NR = [4] will start 1 VM with + 4 interfaces and every NIC pair is directly connected to the + physical ports + * Deployment ``pvvp`` and GUEST_NICS_NR = [2, 4] will start 2 VMs; + 1st VM will have 2 interfaces and 2nd VM 4 interfaces. These interfaces + will be connected in serial, i.e. traffic will flow as follows: + PHY1 -> VM1_1 -> VM1_2 -> VM2_1 -> VM2_2 -> VM2_3 -> VM2_4 -> PHY2 + +Note: In case that only 1 or more than 2 NICs are configured for VM, +then ``testpmd`` should be used as forwarding application inside the VM. +As it is able to forward traffic between multiple VM NIC pairs. + +Note: In case of ``linux_bridge``, all NICs are connected to the same +bridge inside the VM. + +VM, vSwitch, Traffic Generator Independence +=========================================== + +VSPERF supports different vSwithes, Traffic Generators, VNFs +and Forwarding Applications by using standard object-oriented polymorphism: + + * Support for vSwitches is implemented by a class inheriting from IVSwitch. + * Support for Traffic Generators is implemented by a class inheriting from + ITrafficGenerator. + * Support for VNF is implemented by a class inheriting from IVNF. + * Support for Forwarding Applications is implemented by a class inheriting + from IPktFwd. + +By dealing only with the abstract interfaces the core framework can support +many implementations of different vSwitches, Traffic Generators, VNFs +and Forwarding Applications. + +IVSwitch +-------- + +.. code-block:: python + + class IVSwitch: + start(self) + stop(self) + add_switch(switch_name) + del_switch(switch_name) + add_phy_port(switch_name) + add_vport(switch_name) + get_ports(switch_name) + del_port(switch_name, port_name) + add_flow(switch_name, flow) + del_flow(switch_name, flow=None) + +ITrafficGenerator +----------------- + +.. code-block:: python + + class ITrafficGenerator: + connect() + disconnect() + + send_burst_traffic(traffic, numpkts, time, framerate) + + send_cont_traffic(traffic, time, framerate) + start_cont_traffic(traffic, time, framerate) + stop_cont_traffic(self): + + send_rfc2544_throughput(traffic, tests, duration, lossrate) + start_rfc2544_throughput(traffic, tests, duration, lossrate) + wait_rfc2544_throughput(self) + + send_rfc2544_back2back(traffic, tests, duration, lossrate) + start_rfc2544_back2back(traffic, , tests, duration, lossrate) + wait_rfc2544_back2back() + +Note ``send_xxx()`` blocks whereas ``start_xxx()`` does not and must be followed by a subsequent call to ``wait_xxx()``. + +IVnf +---- + +.. code-block:: python + + class IVnf: + start(memory, cpus, + monitor_path, shared_path_host, + shared_path_guest, guest_prompt) + stop() + execute(command) + wait(guest_prompt) + execute_and_wait (command) + +IPktFwd +-------- + + .. code-block:: python + + class IPktFwd: + start() + stop() + + +Controllers +----------- + +Controllers are used in conjunction with abstract interfaces as way +of decoupling the control of vSwtiches, VNFs, TrafficGenerators +and Forwarding Applications from other components. + +The controlled classes provide basic primitive operations. The Controllers +sequence and co-ordinate these primitive operation in to useful actions. For +instance the vswitch_controller_p2p can be used to bring any vSwitch (that +implements the primitives defined in IVSwitch) into the configuration required +by the Phy-to-Phy Deployment Scenario. + +In order to support a new vSwitch only a new implementation of IVSwitch needs +be created for the new vSwitch to be capable of fulfilling all the Deployment +Scenarios provided for by existing or future vSwitch Controllers. + +Similarly if a new Deployment Scenario is required it only needs to be written +once as a new vSwitch Controller and it will immediately be capable of +controlling all existing and future vSwitches in to that Deployment Scenario. + +Similarly the Traffic Controllers can be used to co-ordinate basic operations +provided by implementers of ITrafficGenerator to provide useful tests. Though +traffic generators generally already implement full test cases i.e. they both +generate suitable traffic and analyse returned traffic in order to implement a +test which has typically been predefined in an RFC document. However the +Traffic Controller class allows for the possibility of further enhancement - +such as iterating over tests for various packet sizes or creating new tests. + +Traffic Controller's Role +------------------------- + +.. image:: traffic_controller.png + + +Loader & Component Factory +-------------------------- + +The working of the Loader package (which is responsible for *finding* arbitrary +classes based on configuration data) and the Component Factory which is +responsible for *choosing* the correct class for a particular situation - e.g. +Deployment Scenario can be seen in this diagram. + +.. image:: factory_and_loader.png + +Routing Tables +============== + +Vsperf uses a standard set of routing tables in order to allow tests to easily +mix and match Deployment Scenarios (PVP, P2P topology), Tuple Matching and +Frame Modification requirements. + +.. code-block:: console + + +--------------+ + | | + | Table 0 | table#0 - Match table. Flows designed to force 5 & 10 + | | tuple matches go here. + | | + +--------------+ + | + | + v + +--------------+ table#1 - Routing table. Flow entries to forward + | | packets between ports goes here. + | Table 1 | The chosen port is communicated to subsequent tables by + | | setting the metadata value to the egress port number. + | | Generally this table is set-up by by the + +--------------+ vSwitchController. + | + | + v + +--------------+ table#2 - Frame modification table. Frame modification + | | flow rules are isolated in this table so that they can + | Table 2 | be turned on or off without affecting the routing or + | | tuple-matching flow rules. This allows the frame + | | modification and tuple matching required by the tests + | | in the VSWITCH PERFORMANCE FOR TELCO NFV test + +--------------+ specification to be independent of the Deployment + | Scenario set up by the vSwitchController. + | + v + +--------------+ + | | + | Table 3 | table#3 - Egress table. Egress packets on the ports + | | setup in Table 1. + +--------------+ + + diff --git a/docs/testing/developer/requirements/LICENSE b/docs/testing/developer/requirements/LICENSE new file mode 100644 index 00000000..7bc572ce --- /dev/null +++ b/docs/testing/developer/requirements/LICENSE @@ -0,0 +1,2 @@ +This work is licensed under a Creative Commons Attribution 4.0 International License. +http://creativecommons.org/licenses/by/4.0 diff --git a/docs/testing/developer/requirements/ietf_draft/LICENSE b/docs/testing/developer/requirements/ietf_draft/LICENSE new file mode 100644 index 00000000..7fc9ae14 --- /dev/null +++ b/docs/testing/developer/requirements/ietf_draft/LICENSE @@ -0,0 +1,12 @@ +Copyright (c) 2016 IETF Trust and the persons identified as the +document authors. All rights reserved. + +This document is subject to BCP 78 and the IETF Trust's Legal +Provisions Relating to IETF Documents +(http://trustee.ietf.org/license-info) in effect on the date of +publication of this document. Please review these documents +carefully, as they describe your rights and restrictions with respect +to this document. Code Components extracted from this document must +include Simplified BSD License text as described in Section 4.e of +the Trust Legal Provisions and are provided without warranty as +described in the Simplified BSD License. diff --git a/docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-00.xml b/docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-00.xml new file mode 100644 index 00000000..2259b23c --- /dev/null +++ b/docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-00.xml @@ -0,0 +1,1016 @@ + + + + + + + + + + + + + + + Benchmarking Virtual Switches in + OPNFV + + + Intel + +
+ + + + + + + + + + + + + + + + + maryam.tahhan@intel.com + + +
+ + + + Intel + +
+ + + + + + + + + + + + + + + + + billy.o.mahony@intel.com + + +
+ + + + AT&T Labs + +
+ + 200 Laurel Avenue South + + Middletown, + + NJ + + 07748 + + USA + + + +1 732 420 1571 + + +1 732 368 1192 + + acmorton@att.com + + http://home.comcast.net/~acmacm/ +
+ + + + + + This memo describes the progress of the Open Platform for NFV (OPNFV) + project on virtual switch performance "VSWITCHPERF". This project + intends to build on the current and completed work of the Benchmarking + Methodology Working Group in IETF, by referencing existing literature. + The Benchmarking Methodology Working Group has traditionally conducted + laboratory characterization of dedicated physical implementations of + internetworking functions. Therefore, this memo begins to describe the + additional considerations when virtual switches are implemented in + general-purpose hardware. The expanded tests and benchmarks are also + influenced by the OPNFV mission to support virtualization of the "telco" + infrastructure. + + + + The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", + "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this + document are to be interpreted as described in RFC 2119. + + + + + + +
+ Benchmarking Methodology Working Group (BMWG) has traditionally + conducted laboratory characterization of dedicated physical + implementations of internetworking functions. The Black-box Benchmarks + of Throughput, Latency, Forwarding Rates and others have served our + industry for many years. Now, Network Function Virtualization (NFV) has + the goal to transform how internetwork functions are implemented, and + therefore has garnered much attention. + + This memo summarizes the progress of the Open Platform for NFV + (OPNFV) project on virtual switch performance characterization, + "VSWITCHPERF", through the Brahmaputra (second) release . This project intends to build on the current and + completed work of the Benchmarking Methodology Working Group in IETF, by + referencing existing literature. For example, currently the most often + referenced RFC is (which depends on ) and foundation of the benchmarking work in OPNFV is + common and strong. + + See + https://wiki.opnfv.org/characterize_vswitch_performance_for_telco_nfv_use_cases + for more background, and the OPNFV website for general information: + https://www.opnfv.org/ + + The authors note that OPNFV distinguishes itself from other open + source compute and networking projects through its emphasis on existing + "telco" services as opposed to cloud-computing. There are many ways in + which telco requirements have different emphasis on performance + dimensions when compared to cloud computing: support for and transfer of + isochronous media streams is one example. + + Note also that the move to NFV Infrastructure has resulted in many + new benchmarking initiatives across the industry. The authors are + currently doing their best to maintain alignment with many other + projects, and this Internet Draft is one part of the efforts. We + acknowledge the early work in , and useful + discussion with the authors. +
+ +
+ The primary purpose and scope of the memo is to inform the industry + of work-in-progress that builds on the body of extensive BMWG literature + and experience, and describe the extensions needed for benchmarking + virtual switches. Inital feedback indicates that many of these + extensions may be applicable beyond the current scope (to hardware + switches in the NFV Infrastructure and to virtual routers, for example). + Additionally, this memo serves as a vehicle to include more detail and + commentary from BMWG and other Open Source communities, under BMWG's + chartered work to characterize the NFV Infrastructure (a virtual switch + is an important aspect of that infrastructure). +
+ +
+ This section highlights some specific considerations (from )related to Benchmarks for virtual + switches. The OPNFV project is sharing its present view on these areas, + as they develop their specifications in the Level Test Design (LTD) + document. + +
+ To compare the performance of virtual designs and implementations + with their physical counterparts, identical benchmarks are needed. + BMWG has developed specifications for many network functions this memo + re-uses existing benchmarks through references, and expands them + during development of new methods. A key configuration aspect is the + number of parallel cores required to achieve comparable performance + with a given physical device, or whether some limit of scale was + reached before the cores could achieve the comparable level. + + It's unlikely that the virtual switch will be the only application + running on the SUT, so CPU utilization, Cache utilization, and Memory + footprint should also be recorded for the virtual implementations of + internetworking functions. +
+ +
+ External observations remain essential as the basis for Benchmarks. + Internal observations with fixed specification and interpretation will + be provided in parallel to assist the development of operations + procedures when the technology is deployed. +
+ +
+ A key consideration when conducting any sort of benchmark is trying + to ensure the consistency and repeatability of test results. When + benchmarking the performance of a vSwitch there are many factors that + can affect the consistency of results, one key factor is matching the + various hardware and software details of the SUT. This section lists + some of the many new parameters which this project believes are + critical to report in order to achieve repeatability. + + Hardware details including: + + + Platform details + + Processor details + + Memory information (type and size) + + Number of enabled cores + + Number of cores used for the test + + Number of physical NICs, as well as their details + (manufacturer, versions, type and the PCI slot they are plugged + into) + + NIC interrupt configuration + + BIOS version, release date and any configurations that were + modified + + CPU microcode level + + Memory DIMM configurations (quad rank performance may not be + the same as dual rank) in size, freq and slot locations + + PCI configuration parameters (payload size, early ack + option...) + + Power management at all levels (ACPI sleep states, processor + package, OS...) + Software details including: + + + OS parameters and behavior (text vs graphical no one typing at + the console on one system) + + OS version (for host and VNF) + + Kernel version (for host and VNF) + + GRUB boot parameters (for host and VNF) + + Hypervisor details (Type and version) + + Selected vSwitch, version number or commit id used + + vSwitch launch command line if it has been parameterised + + Memory allocation to the vSwitch + + which NUMA node it is using, and how many memory channels + + DPDK or any other SW dependency version number or commit id + used + + Memory allocation to a VM - if it's from Hugpages/elsewhere + + VM storage type: snapshot/independent persistent/independent + non-persistent + + Number of VMs + + Number of Virtual NICs (vNICs), versions, type and driver + + Number of virtual CPUs and their core affinity on the host + + Number vNIC interrupt configuration + + Thread affinitization for the applications (including the + vSwitch itself) on the host + + Details of Resource isolation, such as CPUs designated for + Host/Kernel (isolcpu) and CPUs designated for specific processes + (taskset). - Test duration. - Number of flows. + + + Test Traffic Information: + Traffic type - UDP, TCP, IMIX / Other + + Packet Sizes + + Deployment Scenario + + + +
+ +
+ Virtual switches group packets into flows by processing and + matching particular packet or frame header information, or by matching + packets based on the input ports. Thus a flow can be thought of a + sequence of packets that have the same set of header field values + (5-tuple) or have arrived on the same port. Performance results can + vary based on the parameters the vSwitch uses to match for a flow. The + recommended flow classification parameters for any vSwitch performance + tests are: the input port, the source IP address, the destination IP + address and the Ethernet protocol type field. It is essential to + increase the flow timeout time on a vSwitch before conducting any + performance tests that do not measure the flow setup time. Normally + the first packet of a particular stream will install the flow in the + virtual switch which adds an additional latency, subsequent packets of + the same flow are not subject to this latency if the flow is already + installed on the vSwitch. +
+ +
+ This outline describes measurement of baseline with isolated + resources at a high level, which is the intended approach at this + time. + + + Baselines: + Optional: Benchmark platform forwarding capability without + a vswitch or VNF for at least 72 hours (serves as a means of + platform validation and a means to obtain the base performance + for the platform in terms of its maximum forwarding rate and + latency).
+ Benchmark platform forwarding + capability + + + + +
+ + Benchmark VNF forwarding capability with direct + connectivity (vSwitch bypass, e.g., SR/IOV) for at least 72 + hours (serves as a means of VNF validation and a means to + obtain the base performance for the VNF in terms of its + maximum forwarding rate and latency). The metrics gathered + from this test will serve as a key comparison point for + vSwitch bypass technologies performance and vSwitch + performance.
+ Benchmark VNF forwarding capability + + + + +
+ + Benchmarking with isolated resources alone, with other + resources (both HW&SW) disabled Example, vSw and VM are + SUT + + Benchmarking with isolated resources alone, leaving some + resources unused + + Benchmark with isolated resources and all resources + occupied + + + Next Steps + Limited sharing + + Production scenarios + + Stressful scenarios + + +
+
+ +
+ The overall specification in preparation is referred to as a Level + Test Design (LTD) document, which will contain a suite of performance + tests. The base performance tests in the LTD are based on the + pre-existing specifications developed by BMWG to test the performance of + physical switches. These specifications include: + + + Benchmarking Methodology for Network + Interconnect Devices + + Benchmarking Methodology for LAN + Switching + + Device Reset Characterization + + Packet Delay Variation Applicability + Statement + + + Some of the above/newer RFCs are being applied in benchmarking for + the first time, and represent a development challenge for test equipment + developers. Fortunately, many members of the testing system community + have engaged on the VSPERF project, including an open source test + system. + + In addition to this, the LTD also re-uses the terminology defined + by: + + + Benchmarking Terminology for LAN + Switching Devices + + Packet Delay Variation Applicability + Statement + + + + + Specifications to be included in future updates of the LTD + include: + Methodology for IP Multicast + Benchmarking + + Packet Reordering Metrics + + + As one might expect, the most fundamental internetworking + characteristics of Throughput and Latency remain important when the + switch is virtualized, and these benchmarks figure prominently in the + specification. + + When considering characteristics important to "telco" network + functions, we must begin to consider additional performance metrics. In + this case, the project specifications have referenced metrics from the + IETF IP Performance Metrics (IPPM) literature. This means that the test of Latency is replaced by measurement of a + metric derived from IPPM's , where a set of + statistical summaries will be provided (mean, max, min, etc.). Further + metrics planned to be benchmarked include packet delay variation as + defined by , reordering, burst behaviour, DUT + availability, DUT capacity and packet loss in long term testing at + Throughput level, where some low-level of background loss may be present + and characterized. + + Tests have been (or will be) designed to collect the metrics + below: + + + Throughput Tests to measure the maximum forwarding rate (in + frames per second or fps) and bit rate (in Mbps) for a constant load + (as defined by ) without traffic loss. + + Packet and Frame Delay Distribution Tests to measure average, min + and max packet and frame delay for constant loads. + + Packet Delay Tests to understand latency distribution for + different packet sizes and over an extended test run to uncover + outliers. + + Scalability Tests to understand how the virtual switch performs + as the number of flows, active ports, complexity of the forwarding + logic’s configuration… it has to deal with + increases. + + Stream Performance Tests (TCP, UDP) to measure bulk data transfer + performance, i.e. how fast systems can send and receive data through + the switch. + + Control Path and Datapath Coupling Tests, to understand how + closely coupled the datapath and the control path are as well as the + effect of this coupling on the performance of the DUT (example: + delay of the initial packet of a flow). + + CPU and Memory Consumption Tests to understand the virtual + switch’s footprint on the system, usually conducted as + auxiliary measurements with benchmarks above. They include: CPU + utilization, Cache utilization and Memory footprint. + + The so-called "Soak" tests, where the selected test is conducted + over a long period of time (with an ideal duration of 24 hours, and + at least 6 hours). The purpose of soak tests is to capture transient + changes in performance which may occur due to infrequent processes + or the low probability coincidence of two or more processes. The + performance must be evaluated periodically during continuous + testing, and this results in use of Frame + Rate metrics instead of Throughput (which + requires stopping traffic to allow time for all traffic to exit + internal queues). + + + Future/planned test specs include: + Request/Response Performance Tests (TCP, UDP) which measure the + transaction rate through the switch. + + Noisy Neighbour Tests, to understand the effects of resource + sharing on the performance of a virtual switch. + + Tests derived from examination of ETSI NFV Draft GS IFA003 + requirements on characterization of + acceleration technologies applied to vswitches. + The flexibility of deployment of a virtual switch within a + network means that the BMWG IETF existing literature needs to be used to + characterize the performance of a switch in various deployment + scenarios. The deployment scenarios under consideration include: + +
+ Physical port to virtual switch to physical + port + + +
+ +
+ Physical port to virtual switch to VNF to virtual switch + to physical port + + +
+ Physical port to virtual switch to VNF to virtual switch + to VNF to virtual switch to physical port + + +
+ Physical port to virtual switch to VNF + + +
+ VNF to virtual switch to physical port + + +
+ VNF to virtual switch to VNF + + +
+ + A set of Deployment Scenario figures is available on the VSPERF Test + Methodology Wiki page . +
+ +
+ This section organizes the many existing test specifications into the + "3x3" matrix (introduced in ). + Because the LTD specification ID names are quite long, this section is + organized into lists for each occupied cell of the matrix (not all are + occupied, also the matrix has grown to 3x4 to accommodate scale metrics + when displaying the coverage of many metrics/benchmarks). The current + version of the LTD specification is available . + + The tests listed below assess the activation of paths in the data + plane, rather than the control plane. + + A complete list of tests with short summaries is available on the + VSPERF "LTD Test Spec Overview" Wiki page . + +
+ + Activation.RFC2889.AddressLearningRate + + PacketLatency.InitialPacketProcessingLatency + +
+ +
+ + CPDP.Coupling.Flow.Addition + +
+ +
+ + Throughput.RFC2544.SystemRecoveryTime + + Throughput.RFC2544.ResetTime + +
+ +
+ + Activation.RFC2889.AddressCachingCapacity + +
+ +
+ + Throughput.RFC2544.PacketLossRate + + CPU.RFC2544.0PacketLoss + + Throughput.RFC2544.PacketLossRateFrameModification + + Throughput.RFC2544.BackToBackFrames + + Throughput.RFC2889.MaxForwardingRate + + Throughput.RFC2889.ForwardPressure + + Throughput.RFC2889.BroadcastFrameForwarding + +
+ +
+ + Throughput.RFC2889.ErrorFramesFiltering + + Throughput.RFC2544.Profile + +
+ +
+ + Throughput.RFC2889.Soak + + Throughput.RFC2889.SoakFrameModification + + PacketDelayVariation.RFC3393.Soak + +
+ +
+ + Scalability.RFC2544.0PacketLoss + + MemoryBandwidth.RFC2544.0PacketLoss.Scalability + +
+ +
+
+ +
+
+
+ +
+ Benchmarking activities as described in this memo are limited to + technology characterization of a Device Under Test/System Under Test + (DUT/SUT) using controlled stimuli in a laboratory environment, with + dedicated address space and the constraints specified in the sections + above. + + The benchmarking network topology will be an independent test setup + and MUST NOT be connected to devices that may forward the test traffic + into a production network, or misroute traffic to the test management + network. + + Further, benchmarking is performed on a "black-box" basis, relying + solely on measurements observable external to the DUT/SUT. + + Special capabilities SHOULD NOT exist in the DUT/SUT specifically for + benchmarking purposes. Any implications for network security arising + from the DUT/SUT SHOULD be identical in the lab and in production + networks. +
+ +
+ No IANA Action is requested at this time. +
+ +
+ The authors appreciate and acknowledge comments from Scott Bradner, + Marius Georgescu, Ramki Krishnan, Doug Montgomery, Martin Klozik, + Christian Trautman, and others for their reviews. +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Network Function Virtualization: Performance and Portability + Best Practices + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Test Topologies + https://wiki.opnfv.org/vsperf/test_methodology + + + + + + + + + + + + LTD Test Spec Overview + https://wiki.opnfv.org/wiki/vswitchperf_test_spec_review + + + + + + + + + + + + LTD Test Specification + http://artifacts.opnfv.org/vswitchperf/docs/requirements/index.html + + + + + + + + + + + + Brahmaputra, Second OPNFV Release + https://www.opnfv.org/brahmaputra + + + + + + + + + + + + https://docbox.etsi.org/ISG/NFV/Open/Drafts/IFA003_Acceleration_-_vSwitch_Spec/ + + + + + + + + + + + diff --git a/docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-01.xml b/docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-01.xml new file mode 100644 index 00000000..c8a3d99b --- /dev/null +++ b/docs/testing/developer/requirements/ietf_draft/draft-ietf-bmwg-vswitch-opnfv-01.xml @@ -0,0 +1,1027 @@ + + + + + + + + + + + + + + + Benchmarking Virtual Switches in + OPNFV + + + Intel + +
+ + + + + + + + + + + + + + + + + maryam.tahhan@intel.com + + +
+ + + + Intel + +
+ + + + + + + + + + + + + + + + + billy.o.mahony@intel.com + + +
+ + + + AT&T Labs + +
+ + 200 Laurel Avenue South + + Middletown, + + NJ + + 07748 + + USA + + + +1 732 420 1571 + + +1 732 368 1192 + + acmorton@att.com + + http://home.comcast.net/~acmacm/ +
+ + + + + + This memo describes the progress of the Open Platform for NFV (OPNFV) + project on virtual switch performance "VSWITCHPERF". This project + intends to build on the current and completed work of the Benchmarking + Methodology Working Group in IETF, by referencing existing literature. + The Benchmarking Methodology Working Group has traditionally conducted + laboratory characterization of dedicated physical implementations of + internetworking functions. Therefore, this memo begins to describe the + additional considerations when virtual switches are implemented in + general-purpose hardware. The expanded tests and benchmarks are also + influenced by the OPNFV mission to support virtualization of the "telco" + infrastructure. + + + + The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", + "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this + document are to be interpreted as described in RFC 2119. + + + + + + +
+ Benchmarking Methodology Working Group (BMWG) has traditionally + conducted laboratory characterization of dedicated physical + implementations of internetworking functions. The Black-box Benchmarks + of Throughput, Latency, Forwarding Rates and others have served our + industry for many years. Now, Network Function Virtualization (NFV) has + the goal to transform how internetwork functions are implemented, and + therefore has garnered much attention. + + This memo summarizes the progress of the Open Platform for NFV + (OPNFV) project on virtual switch performance characterization, + "VSWITCHPERF", through the Brahmaputra (second) release . This project intends to build on the current and + completed work of the Benchmarking Methodology Working Group in IETF, by + referencing existing literature. For example, currently the most often + referenced RFC is (which depends on ) and foundation of the benchmarking work in OPNFV is + common and strong. + + See + https://wiki.opnfv.org/characterize_vswitch_performance_for_telco_nfv_use_cases + for more background, and the OPNFV website for general information: + https://www.opnfv.org/ + + The authors note that OPNFV distinguishes itself from other open + source compute and networking projects through its emphasis on existing + "telco" services as opposed to cloud-computing. There are many ways in + which telco requirements have different emphasis on performance + dimensions when compared to cloud computing: support for and transfer of + isochronous media streams is one example. + + Note also that the move to NFV Infrastructure has resulted in many + new benchmarking initiatives across the industry. The authors are + currently doing their best to maintain alignment with many other + projects, and this Internet Draft is one part of the efforts. We + acknowledge the early work in , and useful + discussion with the authors. +
+ +
+ The primary purpose and scope of the memo is to inform the industry + of work-in-progress that builds on the body of extensive BMWG literature + and experience, and describe the extensions needed for benchmarking + virtual switches. Inital feedback indicates that many of these + extensions may be applicable beyond the current scope (to hardware + switches in the NFV Infrastructure and to virtual routers, for example). + Additionally, this memo serves as a vehicle to include more detail and + commentary from BMWG and other Open Source communities, under BMWG's + chartered work to characterize the NFV Infrastructure (a virtual switch + is an important aspect of that infrastructure). + + The benchmarking covered in this memo should be applicable to many + types of vswitches, and remain vswitch-agnostic to great degree. There + has been no attempt to track and test all features of any specific + vswitch implementation. +
+ +
+ This section highlights some specific considerations (from )related to Benchmarks for virtual + switches. The OPNFV project is sharing its present view on these areas, + as they develop their specifications in the Level Test Design (LTD) + document. + +
+ To compare the performance of virtual designs and implementations + with their physical counterparts, identical benchmarks are needed. + BMWG has developed specifications for many network functions this memo + re-uses existing benchmarks through references, and expands them + during development of new methods. A key configuration aspect is the + number of parallel cores required to achieve comparable performance + with a given physical device, or whether some limit of scale was + reached before the cores could achieve the comparable level. + + It's unlikely that the virtual switch will be the only application + running on the SUT, so CPU utilization, Cache utilization, and Memory + footprint should also be recorded for the virtual implementations of + internetworking functions. +
+ +
+ External observations remain essential as the basis for Benchmarks. + Internal observations with fixed specification and interpretation will + be provided in parallel to assist the development of operations + procedures when the technology is deployed. +
+ +
+ A key consideration when conducting any sort of benchmark is trying + to ensure the consistency and repeatability of test results. When + benchmarking the performance of a vSwitch there are many factors that + can affect the consistency of results, one key factor is matching the + various hardware and software details of the SUT. This section lists + some of the many new parameters which this project believes are + critical to report in order to achieve repeatability. + + Hardware details including: + + + Platform details + + Processor details + + Memory information (type and size) + + Number of enabled cores + + Number of cores used for the test + + Number of physical NICs, as well as their details + (manufacturer, versions, type and the PCI slot they are plugged + into) + + NIC interrupt configuration + + BIOS version, release date and any configurations that were + modified + + CPU microcode level + + Memory DIMM configurations (quad rank performance may not be + the same as dual rank) in size, freq and slot locations + + PCI configuration parameters (payload size, early ack + option...) + + Power management at all levels (ACPI sleep states, processor + package, OS...) + Software details including: + + + OS parameters and behavior (text vs graphical no one typing at + the console on one system) + + OS version (for host and VNF) + + Kernel version (for host and VNF) + + GRUB boot parameters (for host and VNF) + + Hypervisor details (Type and version) + + Selected vSwitch, version number or commit id used + + vSwitch launch command line if it has been parameterised + + Memory allocation to the vSwitch + + which NUMA node it is using, and how many memory channels + + DPDK or any other SW dependency version number or commit id + used + + Memory allocation to a VM - if it's from Hugpages/elsewhere + + VM storage type: snapshot/independent persistent/independent + non-persistent + + Number of VMs + + Number of Virtual NICs (vNICs), versions, type and driver + + Number of virtual CPUs and their core affinity on the host + + Number vNIC interrupt configuration + + Thread affinitization for the applications (including the + vSwitch itself) on the host + + Details of Resource isolation, such as CPUs designated for + Host/Kernel (isolcpu) and CPUs designated for specific processes + (taskset). - Test duration. - Number of flows. + + + Test Traffic Information: + Traffic type - UDP, TCP, IMIX / Other + + Packet Sizes + + Deployment Scenario + + + +
+ +
+ Virtual switches group packets into flows by processing and + matching particular packet or frame header information, or by matching + packets based on the input ports. Thus a flow can be thought of a + sequence of packets that have the same set of header field values + (5-tuple) or have arrived on the same port. Performance results can + vary based on the parameters the vSwitch uses to match for a flow. The + recommended flow classification parameters for any vSwitch performance + tests are: the input port, the source IP address, the destination IP + address and the Ethernet protocol type field. It is essential to + increase the flow timeout time on a vSwitch before conducting any + performance tests that do not measure the flow setup time. Normally + the first packet of a particular stream will install the flow in the + virtual switch which adds an additional latency, subsequent packets of + the same flow are not subject to this latency if the flow is already + installed on the vSwitch. +
+ +
+ This outline describes measurement of baseline with isolated + resources at a high level, which is the intended approach at this + time. + + + Baselines: + Optional: Benchmark platform forwarding capability without + a vswitch or VNF for at least 72 hours (serves as a means of + platform validation and a means to obtain the base performance + for the platform in terms of its maximum forwarding rate and + latency).
+ Benchmark platform forwarding + capability + + + + +
+ + Benchmark VNF forwarding capability with direct + connectivity (vSwitch bypass, e.g., SR/IOV) for at least 72 + hours (serves as a means of VNF validation and a means to + obtain the base performance for the VNF in terms of its + maximum forwarding rate and latency). The metrics gathered + from this test will serve as a key comparison point for + vSwitch bypass technologies performance and vSwitch + performance.
+ Benchmark VNF forwarding capability + + + + +
+ + Benchmarking with isolated resources alone, with other + resources (both HW&SW) disabled Example, vSw and VM are + SUT + + Benchmarking with isolated resources alone, leaving some + resources unused + + Benchmark with isolated resources and all resources + occupied + + + Next Steps + Limited sharing + + Production scenarios + + Stressful scenarios + + +
+
+ +
+ The overall specification in preparation is referred to as a Level + Test Design (LTD) document, which will contain a suite of performance + tests. The base performance tests in the LTD are based on the + pre-existing specifications developed by BMWG to test the performance of + physical switches. These specifications include: + + + Benchmarking Methodology for Network + Interconnect Devices + + Benchmarking Methodology for LAN + Switching + + Device Reset Characterization + + Packet Delay Variation Applicability + Statement + + + Some of the above/newer RFCs are being applied in benchmarking for + the first time, and represent a development challenge for test equipment + developers. Fortunately, many members of the testing system community + have engaged on the VSPERF project, including an open source test + system. + + In addition to this, the LTD also re-uses the terminology defined + by: + + + Benchmarking Terminology for LAN + Switching Devices + + Packet Delay Variation Applicability + Statement + + + + + Specifications to be included in future updates of the LTD + include: + Methodology for IP Multicast + Benchmarking + + Packet Reordering Metrics + + + As one might expect, the most fundamental internetworking + characteristics of Throughput and Latency remain important when the + switch is virtualized, and these benchmarks figure prominently in the + specification. + + When considering characteristics important to "telco" network + functions, we must begin to consider additional performance metrics. In + this case, the project specifications have referenced metrics from the + IETF IP Performance Metrics (IPPM) literature. This means that the test of Latency is replaced by measurement of a + metric derived from IPPM's , where a set of + statistical summaries will be provided (mean, max, min, etc.). Further + metrics planned to be benchmarked include packet delay variation as + defined by , reordering, burst behaviour, DUT + availability, DUT capacity and packet loss in long term testing at + Throughput level, where some low-level of background loss may be present + and characterized. + + Tests have been (or will be) designed to collect the metrics + below: + + + Throughput Tests to measure the maximum forwarding rate (in + frames per second or fps) and bit rate (in Mbps) for a constant load + (as defined by ) without traffic loss. + + Packet and Frame Delay Distribution Tests to measure average, min + and max packet and frame delay for constant loads. + + Packet Delay Tests to understand latency distribution for + different packet sizes and over an extended test run to uncover + outliers. + + Scalability Tests to understand how the virtual switch performs + as the number of flows, active ports, complexity of the forwarding + logic’s configuration… it has to deal with + increases. + + Stream Performance Tests (TCP, UDP) to measure bulk data transfer + performance, i.e. how fast systems can send and receive data through + the switch. + + Control Path and Datapath Coupling Tests, to understand how + closely coupled the datapath and the control path are as well as the + effect of this coupling on the performance of the DUT (example: + delay of the initial packet of a flow). + + CPU and Memory Consumption Tests to understand the virtual + switch’s footprint on the system, usually conducted as + auxiliary measurements with benchmarks above. They include: CPU + utilization, Cache utilization and Memory footprint. + + The so-called "Soak" tests, where the selected test is conducted + over a long period of time (with an ideal duration of 24 hours, but + only long enough to determine that stability issues exist when + found; there is no requirement to continue a test when a DUT + exhibits instability over time). The key performance characteristics + and benchmarks for a DUT are determined (using short duration tests) + prior to conducting soak tests. The purpose of soak tests is to + capture transient changes in performance which may occur due to + infrequent processes, memory leaks, or the low probability + coincidence of two or more processes. The stability of the DUT is + the paramount consideration, so performance must be evaluated + periodically during continuous testing, and this results in use of + Frame Rate metrics instead of Throughput (which requires stopping traffic to + allow time for all traffic to exit internal queues), for + example. + + + Future/planned test specs include: + Request/Response Performance Tests (TCP, UDP) which measure the + transaction rate through the switch. + + Noisy Neighbour Tests, to understand the effects of resource + sharing on the performance of a virtual switch. + + Tests derived from examination of ETSI NFV Draft GS IFA003 + requirements on characterization of + acceleration technologies applied to vswitches. + The flexibility of deployment of a virtual switch within a + network means that the BMWG IETF existing literature needs to be used to + characterize the performance of a switch in various deployment + scenarios. The deployment scenarios under consideration include: + +
+ Physical port to virtual switch to physical + port + + +
+ +
+ Physical port to virtual switch to VNF to virtual switch + to physical port + + +
+ Physical port to virtual switch to VNF to virtual switch + to VNF to virtual switch to physical port + + +
+ Physical port to virtual switch to VNF + + +
+ VNF to virtual switch to physical port + + +
+ VNF to virtual switch to VNF + + +
+ + A set of Deployment Scenario figures is available on the VSPERF Test + Methodology Wiki page . +
+ +
+ This section organizes the many existing test specifications into the + "3x3" matrix (introduced in ). + Because the LTD specification ID names are quite long, this section is + organized into lists for each occupied cell of the matrix (not all are + occupied, also the matrix has grown to 3x4 to accommodate scale metrics + when displaying the coverage of many metrics/benchmarks). The current + version of the LTD specification is available . + + The tests listed below assess the activation of paths in the data + plane, rather than the control plane. + + A complete list of tests with short summaries is available on the + VSPERF "LTD Test Spec Overview" Wiki page . + +
+ + Activation.RFC2889.AddressLearningRate + + PacketLatency.InitialPacketProcessingLatency + +
+ +
+ + CPDP.Coupling.Flow.Addition + +
+ +
+ + Throughput.RFC2544.SystemRecoveryTime + + Throughput.RFC2544.ResetTime + +
+ +
+ + Activation.RFC2889.AddressCachingCapacity + +
+ +
+ + Throughput.RFC2544.PacketLossRate + + CPU.RFC2544.0PacketLoss + + Throughput.RFC2544.PacketLossRateFrameModification + + Throughput.RFC2544.BackToBackFrames + + Throughput.RFC2889.MaxForwardingRate + + Throughput.RFC2889.ForwardPressure + + Throughput.RFC2889.BroadcastFrameForwarding + +
+ +
+ + Throughput.RFC2889.ErrorFramesFiltering + + Throughput.RFC2544.Profile + +
+ +
+ + Throughput.RFC2889.Soak + + Throughput.RFC2889.SoakFrameModification + + PacketDelayVariation.RFC3393.Soak + +
+ +
+ + Scalability.RFC2544.0PacketLoss + + MemoryBandwidth.RFC2544.0PacketLoss.Scalability + +
+ +
+
+ +
+
+
+ +
+ Benchmarking activities as described in this memo are limited to + technology characterization of a Device Under Test/System Under Test + (DUT/SUT) using controlled stimuli in a laboratory environment, with + dedicated address space and the constraints specified in the sections + above. + + The benchmarking network topology will be an independent test setup + and MUST NOT be connected to devices that may forward the test traffic + into a production network, or misroute traffic to the test management + network. + + Further, benchmarking is performed on a "black-box" basis, relying + solely on measurements observable external to the DUT/SUT. + + Special capabilities SHOULD NOT exist in the DUT/SUT specifically for + benchmarking purposes. Any implications for network security arising + from the DUT/SUT SHOULD be identical in the lab and in production + networks. +
+ +
+ No IANA Action is requested at this time. +
+ +
+ The authors appreciate and acknowledge comments from Scott Bradner, + Marius Georgescu, Ramki Krishnan, Doug Montgomery, Martin Klozik, + Christian Trautman, and others for their reviews. +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Network Function Virtualization: Performance and Portability + Best Practices + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Test Topologies + https://wiki.opnfv.org/vsperf/test_methodology + + + + + + + + + + + + LTD Test Spec Overview + https://wiki.opnfv.org/wiki/vswitchperf_test_spec_review + + + + + + + + + + + + LTD Test Specification + http://artifacts.opnfv.org/vswitchperf/brahmaputra/docs/requirements/index.html + + + + + + + + + + + + Brahmaputra, Second OPNFV Release + https://www.opnfv.org/brahmaputra + + + + + + + + + + + + https://docbox.etsi.org/ISG/NFV/Open/Drafts/IFA003_Acceleration_-_vSwitch_Spec/ + + + + + + + + + + + diff --git a/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-00.xml b/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-00.xml new file mode 100644 index 00000000..b5f7f833 --- /dev/null +++ b/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-00.xml @@ -0,0 +1,964 @@ + + + + + + + + + + + + + + + Benchmarking Virtual Switches in + OPNFV + + + Intel + +
+ + + + + + + + + + + + + + + + + maryam.tahhan@intel.com + + +
+ + + + Intel + +
+ + + + + + + + + + + + + + + + + billy.o.mahony@intel.com + + +
+ + + + AT&T Labs + +
+ + 200 Laurel Avenue South + + Middletown, + + NJ + + 07748 + + USA + + + +1 732 420 1571 + + +1 732 368 1192 + + acmorton@att.com + + http://home.comcast.net/~acmacm/ +
+ + + + + + This memo describes the progress of the Open Platform for NFV (OPNFV) + project on virtual switch performance "VSWITCHPERF". This project + intends to build on the current and completed work of the Benchmarking + Methodology Working Group in IETF, by referencing existing literature. + The Benchmarking Methodology Working Group has traditionally conducted + laboratory characterization of dedicated physical implementations of + internetworking functions. Therefore, this memo begins to describe the + additional considerations when virtual switches are implemented in + general-purpose hardware. The expanded tests and benchmarks are also + influenced by the OPNFV mission to support virtualization of the "telco" + infrastructure. + + + + The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", + "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this + document are to be interpreted as described in RFC 2119. + + + + + + +
+ Benchmarking Methodology Working Group (BMWG) has traditionally + conducted laboratory characterization of dedicated physical + implementations of internetworking functions. The Black-box Benchmarks + of Throughput, Latency, Forwarding Rates and others have served our + industry for many years. Now, Network Function Virtualization (NFV) has + the goal to transform how internetwork functions are implemented, and + therefore has garnered much attention. + + This memo describes the progress of the Open Platform for NFV (OPNFV) + project on virtual switch performance characterization, "VSWITCHPERF". + This project intends to build on the current and completed work of the + Benchmarking Methodology Working Group in IETF, by referencing existing + literature. For example, currently the most often referenced RFC is + (which depends on ) and + foundation of the benchmarking work in OPNFV is common and strong. + + See + https://wiki.opnfv.org/characterize_vswitch_performance_for_telco_nfv_use_cases + for more background, and the OPNFV website for general information: + https://www.opnfv.org/ + + The authors note that OPNFV distinguishes itself from other open + source compute and networking projects through its emphasis on existing + "telco" services as opposed to cloud-computing. There are many ways in + which telco requirements have different emphasis on performance + dimensions when compared to cloud computing: support for and transfer of + isochronous media streams is one example. + + Note also that the move to NFV Infrastructure has resulted in many + new benchmarking initiatives across the industry, and the authors are + currently doing their best to maintain alignment with many other + projects, and this Internet Draft is evidence of the efforts. +
+ +
+ The primary purpose and scope of the memo is to inform BMWG of + work-in-progress that builds on the body of extensive literature and + experience. Additionally, once the initial information conveyed here is + received, this memo may be expanded to include more detail and + commentary from both BMWG and OPNFV communities, under BMWG's chartered + work to characterize the NFV Infrastructure (a virtual switch is an + important aspect of that infrastructure). +
+ +
+ This section highlights some specific considerations (from )related to Benchmarks for virtual + switches. The OPNFV project is sharing its present view on these areas, + as they develop their specifications in the Level Test Design (LTD) + document. + +
+ To compare the performance of virtual designs and implementations + with their physical counterparts, identical benchmarks are needed. + BMWG has developed specifications for many network functions this memo + re-uses existing benchmarks through references, and expands them + during development of new methods. A key configuration aspect is the + number of parallel cores required to achieve comparable performance + with a given physical device, or whether some limit of scale was + reached before the cores could achieve the comparable level. + + It's unlikely that the virtual switch will be the only application + running on the SUT, so CPU utilization, Cache utilization, and Memory + footprint should also be recorded for the virtual implementations of + internetworking functions. +
+ +
+ External observations remain essential as the basis for Benchmarks. + Internal observations with fixed specification and interpretation will + be provided in parallel to assist the development of operations + procedures when the technology is deployed. +
+ +
+ A key consideration when conducting any sort of benchmark is trying + to ensure the consistency and repeatability of test results. When + benchmarking the performance of a vSwitch there are many factors that + can affect the consistency of results, one key factor is matching the + various hardware and software details of the SUT. This section lists + some of the many new parameters which this project believes are + critical to report in order to achieve repeatability. + + Hardware details including: + + + Platform details + + Processor details + + Memory information (type and size) + + Number of enabled cores + + Number of cores used for the test + + Number of physical NICs, as well as their details + (manufacturer, versions, type and the PCI slot they are plugged + into) + + NIC interrupt configuration + + BIOS version, release date and any configurations that were + modified + + CPU microcode level + + Memory DIMM configurations (quad rank performance may not be + the same as dual rank) in size, freq and slot locations + + PCI configuration parameters (payload size, early ack + option...) + + Power management at all levels (ACPI sleep states, processor + package, OS...) + Software details including: + + + OS parameters and behavior (text vs graphical no one typing at + the console on one system) + + OS version (for host and VNF) + + Kernel version (for host and VNF) + + GRUB boot parameters (for host and VNF) + + Hypervisor details (Type and version) + + Selected vSwitch, version number or commit id used + + vSwitch launch command line if it has been parameterised + + Memory allocation to the vSwitch + + which NUMA node it is using, and how many memory channels + + DPDK or any other SW dependency version number or commit id + used + + Memory allocation to a VM - if it's from Hugpages/elsewhere + + VM storage type: snapshot/independent persistent/independent + non-persistent + + Number of VMs + + Number of Virtual NICs (vNICs), versions, type and driver + + Number of virtual CPUs and their core affinity on the host + + Number vNIC interrupt configuration + + Thread affinitization for the applications (including the + vSwitch itself) on the host + + Details of Resource isolation, such as CPUs designated for + Host/Kernel (isolcpu) and CPUs designated for specific processes + (taskset). - Test duration. - Number of flows. + + + Test Traffic Information: + Traffic type - UDP, TCP, IMIX / Other + + Packet Sizes + + Deployment Scenario + + + +
+ +
+ Virtual switches group packets into flows by processing and + matching particular packet or frame header information, or by matching + packets based on the input ports. Thus a flow can be thought of a + sequence of packets that have the same set of header field values or + have arrived on the same port. Performance results can vary based on + the parameters the vSwitch uses to match for a flow. The recommended + flow classification parameters for any vSwitch performance tests are: + the input port, the source IP address, the destination IP address and + the Ethernet protocol type field. It is essential to increase the flow + timeout time on a vSwitch before conducting any performance tests that + do not measure the flow setup time. Normally the first packet of a + particular stream will install the flow in the virtual switch which + adds an additional latency, subsequent packets of the same flow are + not subject to this latency if the flow is already installed on the + vSwitch. +
+ +
+ This outline describes measurement of baseline with isolated + resources at a high level, which is the intended approach at this + time. + + + Baselines: + Optional: Benchmark platform forwarding capability without + a vswitch or VNF for at least 72 hours (serves as a means of + platform validation and a means to obtain the base performance + for the platform in terms of its maximum forwarding rate and + latency).
+ Benchmark platform forwarding + capability + + + + +
+ + Benchmark VNF forwarding capability with direct + connectivity (vSwitch bypass, e.g., SR/IOV) for at least 72 + hours (serves as a means of VNF validation and a means to + obtain the base performance for the VNF in terms of its + maximum forwarding rate and latency). The metrics gathered + from this test will serve as a key comparison point for + vSwitch bypass technologies performance and vSwitch + performance.
+ Benchmark VNF forwarding capability + + + + +
+ + Benchmarking with isolated resources alone, with other + resources (both HW&SW) disabled Example, vSw and VM are + SUT + + Benchmarking with isolated resources alone, leaving some + resources unused + + Benchmark with isolated resources and all resources + occupied + + + Next Steps + Limited sharing + + Production scenarios + + Stressful scenarios + + +
+
+ +
+ The overall specification in preparation is referred to as a Level + Test Design (LTD) document, which will contain a suite of performance + tests. The base performance tests in the LTD are based on the + pre-existing specifications developed by BMWG to test the performance of + physical switches. These specifications include: + + + Benchmarking Methodology for Network + Interconnect Devices + + Benchmarking Methodology for LAN + Switching + + Device Reset Characterization + + Packet Delay Variation Applicability + Statement + + + Some of the above/newer RFCs are being applied in benchmarking for + the first time, and represent a development challenge for test equipment + developers. Fortunately, many members of the testing system community + have engaged on the VSPERF project, including an open source test + system. + + In addition to this, the LTD also re-uses the terminology defined + by: + + + Benchmarking Terminology for LAN + Switching Devices + + Packet Delay Variation Applicability + Statement + + + + + Specifications to be included in future updates of the LTD + include: + Methodology for IP Multicast + Benchmarking + + Packet Reordering Metrics + + + As one might expect, the most fundamental internetworking + characteristics of Throughput and Latency remain important when the + switch is virtualized, and these benchmarks figure prominently in the + specification. + + When considering characteristics important to "telco" network + functions, we must begin to consider additional performance metrics. In + this case, the project specifications have referenced metrics from the + IETF IP Performance Metrics (IPPM) literature. This means that the test of Latency is replaced by measurement of a + metric derived from IPPM's , where a set of + statistical summaries will be provided (mean, max, min, etc.). Further + metrics planned to be benchmarked include packet delay variation as + defined by , reordering, burst behaviour, DUT + availability, DUT capacity and packet loss in long term testing at + Throughput level, where some low-level of background loss may be present + and characterized. + + Tests have been (or will be) designed to collect the metrics + below: + + + Throughput Tests to measure the maximum forwarding rate (in + frames per second or fps) and bit rate (in Mbps) for a constant load + (as defined by RFC1242) without traffic loss. + + Packet and Frame Delay Distribution Tests to measure average, min + and max packet and frame delay for constant loads. + + Packet Delay Tests to understand latency distribution for + different packet sizes and over an extended test run to uncover + outliers. + + Scalability Tests to understand how the virtual switch performs + as the number of flows, active ports, complexity of the forwarding + logic’s configuration… it has to deal with + increases. + + Stream Performance Tests (TCP, UDP) to measure bulk data transfer + performance, i.e. how fast systems can send and receive data through + the switch. + + Control Path and Datapath Coupling Tests, to understand how + closely coupled the datapath and the control path are as well as the + effect of this coupling on the performance of the DUT (example: + delay of the initial packet of a flow). + + CPU and Memory Consumption Tests to understand the virtual + switch’s footprint on the system, usually conducted as + auxiliary measurements with benchmarks above. They include: CPU + utilization, Cache utilization and Memory footprint. + + + Future/planned test specs include: + Request/Response Performance Tests (TCP, UDP) which measure the + transaction rate through the switch. + + Noisy Neighbour Tests, to understand the effects of resource + sharing on the performance of a virtual switch. + + Tests derived from examination of ETSI NFV Draft GS IFA003 + requirements on characterization of + acceleration technologies applied to vswitches. + The flexibility of deployment of a virtual switch within a + network means that the BMWG IETF existing literature needs to be used to + characterize the performance of a switch in various deployment + scenarios. The deployment scenarios under consideration include: + +
+ Physical port to virtual switch to physical + port + + +
+ +
+ Physical port to virtual switch to VNF to virtual switch + to physical port + + +
+ Physical port to virtual switch to VNF to virtual switch + to VNF to virtual switch to physical port + + +
+ Physical port to virtual switch to VNF + + +
+ VNF to virtual switch to physical port + + +
+ VNF to virtual switch to VNF + + +
+ + A set of Deployment Scenario figures is available on the VSPERF Test + Methodology Wiki page . +
+ +
+ This section organizes the many existing test specifications into the + "3x3" matrix (introduced in ). + Because the LTD specification ID names are quite long, this section is + organized into lists for each occupied cell of the matrix (not all are + occupied, also the matrix has grown to 3x4 to accommodate scale metrics + when displaying the coverage of many metrics/benchmarks). + + The tests listed below assess the activation of paths in the data + plane, rather than the control plane. + + A complete list of tests with short summaries is available on the + VSPERF "LTD Test Spec Overview" Wiki page . + +
+ + Activation.RFC2889.AddressLearningRate + + PacketLatency.InitialPacketProcessingLatency + +
+ +
+ + CPDP.Coupling.Flow.Addition + +
+ +
+ + Throughput.RFC2544.SystemRecoveryTime + + Throughput.RFC2544.ResetTime + +
+ +
+ + Activation.RFC2889.AddressCachingCapacity + +
+ +
+ + Throughput.RFC2544.PacketLossRate + + CPU.RFC2544.0PacketLoss + + Throughput.RFC2544.PacketLossRateFrameModification + + Throughput.RFC2544.BackToBackFrames + + Throughput.RFC2889.MaxForwardingRate + + Throughput.RFC2889.ForwardPressure + + Throughput.RFC2889.BroadcastFrameForwarding + +
+ +
+ + Throughput.RFC2889.ErrorFramesFiltering + + Throughput.RFC2544.Profile + +
+ +
+ + Throughput.RFC2889.Soak + + Throughput.RFC2889.SoakFrameModification + + PacketDelayVariation.RFC3393.Soak + +
+ +
+ + Scalability.RFC2544.0PacketLoss + + MemoryBandwidth.RFC2544.0PacketLoss.Scalability + +
+ +
+
+ +
+
+
+ +
+ Benchmarking activities as described in this memo are limited to + technology characterization of a Device Under Test/System Under Test + (DUT/SUT) using controlled stimuli in a laboratory environment, with + dedicated address space and the constraints specified in the sections + above. + + The benchmarking network topology will be an independent test setup + and MUST NOT be connected to devices that may forward the test traffic + into a production network, or misroute traffic to the test management + network. + + Further, benchmarking is performed on a "black-box" basis, relying + solely on measurements observable external to the DUT/SUT. + + Special capabilities SHOULD NOT exist in the DUT/SUT specifically for + benchmarking purposes. Any implications for network security arising + from the DUT/SUT SHOULD be identical in the lab and in production + networks. +
+ +
+ No IANA Action is requested at this time. +
+ +
+ The authors acknowledge +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Network Function Virtualization: Performance and Portability + Best Practices + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Test Topologies + https://wiki.opnfv.org/vsperf/test_methodology + + + + + + + + + + + + LTD Test Spec Overview + https://wiki.opnfv.org/wiki/vswitchperf_test_spec_review + + + + + + + + + + + + https://docbox.etsi.org/ISG/NFV/Open/Drafts/IFA003_Acceleration_-_vSwitch_Spec/ + + + + + + + + + + + diff --git a/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-01.xml b/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-01.xml new file mode 100644 index 00000000..a9405a77 --- /dev/null +++ b/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-01.xml @@ -0,0 +1,1016 @@ + + + + + + + + + + + + + + + Benchmarking Virtual Switches in + OPNFV + + + Intel + +
+ + + + + + + + + + + + + + + + + maryam.tahhan@intel.com + + +
+ + + + Intel + +
+ + + + + + + + + + + + + + + + + billy.o.mahony@intel.com + + +
+ + + + AT&T Labs + +
+ + 200 Laurel Avenue South + + Middletown, + + NJ + + 07748 + + USA + + + +1 732 420 1571 + + +1 732 368 1192 + + acmorton@att.com + + http://home.comcast.net/~acmacm/ +
+ + + + + + This memo describes the progress of the Open Platform for NFV (OPNFV) + project on virtual switch performance "VSWITCHPERF". This project + intends to build on the current and completed work of the Benchmarking + Methodology Working Group in IETF, by referencing existing literature. + The Benchmarking Methodology Working Group has traditionally conducted + laboratory characterization of dedicated physical implementations of + internetworking functions. Therefore, this memo begins to describe the + additional considerations when virtual switches are implemented in + general-purpose hardware. The expanded tests and benchmarks are also + influenced by the OPNFV mission to support virtualization of the "telco" + infrastructure. + + + + The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", + "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this + document are to be interpreted as described in RFC 2119. + + + + + + +
+ Benchmarking Methodology Working Group (BMWG) has traditionally + conducted laboratory characterization of dedicated physical + implementations of internetworking functions. The Black-box Benchmarks + of Throughput, Latency, Forwarding Rates and others have served our + industry for many years. Now, Network Function Virtualization (NFV) has + the goal to transform how internetwork functions are implemented, and + therefore has garnered much attention. + + This memo summarizes the progress of the Open Platform for NFV + (OPNFV) project on virtual switch performance characterization, + "VSWITCHPERF", through the Brahmaputra (second) release . This project intends to build on the current and + completed work of the Benchmarking Methodology Working Group in IETF, by + referencing existing literature. For example, currently the most often + referenced RFC is (which depends on ) and foundation of the benchmarking work in OPNFV is + common and strong. + + See + https://wiki.opnfv.org/characterize_vswitch_performance_for_telco_nfv_use_cases + for more background, and the OPNFV website for general information: + https://www.opnfv.org/ + + The authors note that OPNFV distinguishes itself from other open + source compute and networking projects through its emphasis on existing + "telco" services as opposed to cloud-computing. There are many ways in + which telco requirements have different emphasis on performance + dimensions when compared to cloud computing: support for and transfer of + isochronous media streams is one example. + + Note also that the move to NFV Infrastructure has resulted in many + new benchmarking initiatives across the industry. The authors are + currently doing their best to maintain alignment with many other + projects, and this Internet Draft is one part of the efforts. We + acknowledge the early work in , and useful + discussion with the authors. +
+ +
+ The primary purpose and scope of the memo is to inform the industry + of work-in-progress that builds on the body of extensive BMWG literature + and experience, and describe the extensions needed for benchmarking + virtual switches. Inital feedback indicates that many of these + extensions may be applicable beyond the current scope (to hardware + switches in the NFV Infrastructure and to virtual routers, for example). + Additionally, this memo serves as a vehicle to include more detail and + commentary from BMWG and other Open Source communities, under BMWG's + chartered work to characterize the NFV Infrastructure (a virtual switch + is an important aspect of that infrastructure). +
+ +
+ This section highlights some specific considerations (from )related to Benchmarks for virtual + switches. The OPNFV project is sharing its present view on these areas, + as they develop their specifications in the Level Test Design (LTD) + document. + +
+ To compare the performance of virtual designs and implementations + with their physical counterparts, identical benchmarks are needed. + BMWG has developed specifications for many network functions this memo + re-uses existing benchmarks through references, and expands them + during development of new methods. A key configuration aspect is the + number of parallel cores required to achieve comparable performance + with a given physical device, or whether some limit of scale was + reached before the cores could achieve the comparable level. + + It's unlikely that the virtual switch will be the only application + running on the SUT, so CPU utilization, Cache utilization, and Memory + footprint should also be recorded for the virtual implementations of + internetworking functions. +
+ +
+ External observations remain essential as the basis for Benchmarks. + Internal observations with fixed specification and interpretation will + be provided in parallel to assist the development of operations + procedures when the technology is deployed. +
+ +
+ A key consideration when conducting any sort of benchmark is trying + to ensure the consistency and repeatability of test results. When + benchmarking the performance of a vSwitch there are many factors that + can affect the consistency of results, one key factor is matching the + various hardware and software details of the SUT. This section lists + some of the many new parameters which this project believes are + critical to report in order to achieve repeatability. + + Hardware details including: + + + Platform details + + Processor details + + Memory information (type and size) + + Number of enabled cores + + Number of cores used for the test + + Number of physical NICs, as well as their details + (manufacturer, versions, type and the PCI slot they are plugged + into) + + NIC interrupt configuration + + BIOS version, release date and any configurations that were + modified + + CPU microcode level + + Memory DIMM configurations (quad rank performance may not be + the same as dual rank) in size, freq and slot locations + + PCI configuration parameters (payload size, early ack + option...) + + Power management at all levels (ACPI sleep states, processor + package, OS...) + Software details including: + + + OS parameters and behavior (text vs graphical no one typing at + the console on one system) + + OS version (for host and VNF) + + Kernel version (for host and VNF) + + GRUB boot parameters (for host and VNF) + + Hypervisor details (Type and version) + + Selected vSwitch, version number or commit id used + + vSwitch launch command line if it has been parameterised + + Memory allocation to the vSwitch + + which NUMA node it is using, and how many memory channels + + DPDK or any other SW dependency version number or commit id + used + + Memory allocation to a VM - if it's from Hugpages/elsewhere + + VM storage type: snapshot/independent persistent/independent + non-persistent + + Number of VMs + + Number of Virtual NICs (vNICs), versions, type and driver + + Number of virtual CPUs and their core affinity on the host + + Number vNIC interrupt configuration + + Thread affinitization for the applications (including the + vSwitch itself) on the host + + Details of Resource isolation, such as CPUs designated for + Host/Kernel (isolcpu) and CPUs designated for specific processes + (taskset). - Test duration. - Number of flows. + + + Test Traffic Information: + Traffic type - UDP, TCP, IMIX / Other + + Packet Sizes + + Deployment Scenario + + + +
+ +
+ Virtual switches group packets into flows by processing and + matching particular packet or frame header information, or by matching + packets based on the input ports. Thus a flow can be thought of a + sequence of packets that have the same set of header field values or + have arrived on the same port. Performance results can vary based on + the parameters the vSwitch uses to match for a flow. The recommended + flow classification parameters for any vSwitch performance tests are: + the input port, the source IP address, the destination IP address and + the Ethernet protocol type field. It is essential to increase the flow + timeout time on a vSwitch before conducting any performance tests that + do not measure the flow setup time. Normally the first packet of a + particular stream will install the flow in the virtual switch which + adds an additional latency, subsequent packets of the same flow are + not subject to this latency if the flow is already installed on the + vSwitch. +
+ +
+ This outline describes measurement of baseline with isolated + resources at a high level, which is the intended approach at this + time. + + + Baselines: + Optional: Benchmark platform forwarding capability without + a vswitch or VNF for at least 72 hours (serves as a means of + platform validation and a means to obtain the base performance + for the platform in terms of its maximum forwarding rate and + latency).
+ Benchmark platform forwarding + capability + + + + +
+ + Benchmark VNF forwarding capability with direct + connectivity (vSwitch bypass, e.g., SR/IOV) for at least 72 + hours (serves as a means of VNF validation and a means to + obtain the base performance for the VNF in terms of its + maximum forwarding rate and latency). The metrics gathered + from this test will serve as a key comparison point for + vSwitch bypass technologies performance and vSwitch + performance.
+ Benchmark VNF forwarding capability + + + + +
+ + Benchmarking with isolated resources alone, with other + resources (both HW&SW) disabled Example, vSw and VM are + SUT + + Benchmarking with isolated resources alone, leaving some + resources unused + + Benchmark with isolated resources and all resources + occupied + + + Next Steps + Limited sharing + + Production scenarios + + Stressful scenarios + + +
+
+ +
+ The overall specification in preparation is referred to as a Level + Test Design (LTD) document, which will contain a suite of performance + tests. The base performance tests in the LTD are based on the + pre-existing specifications developed by BMWG to test the performance of + physical switches. These specifications include: + + + Benchmarking Methodology for Network + Interconnect Devices + + Benchmarking Methodology for LAN + Switching + + Device Reset Characterization + + Packet Delay Variation Applicability + Statement + + + Some of the above/newer RFCs are being applied in benchmarking for + the first time, and represent a development challenge for test equipment + developers. Fortunately, many members of the testing system community + have engaged on the VSPERF project, including an open source test + system. + + In addition to this, the LTD also re-uses the terminology defined + by: + + + Benchmarking Terminology for LAN + Switching Devices + + Packet Delay Variation Applicability + Statement + + + + + Specifications to be included in future updates of the LTD + include: + Methodology for IP Multicast + Benchmarking + + Packet Reordering Metrics + + + As one might expect, the most fundamental internetworking + characteristics of Throughput and Latency remain important when the + switch is virtualized, and these benchmarks figure prominently in the + specification. + + When considering characteristics important to "telco" network + functions, we must begin to consider additional performance metrics. In + this case, the project specifications have referenced metrics from the + IETF IP Performance Metrics (IPPM) literature. This means that the test of Latency is replaced by measurement of a + metric derived from IPPM's , where a set of + statistical summaries will be provided (mean, max, min, etc.). Further + metrics planned to be benchmarked include packet delay variation as + defined by , reordering, burst behaviour, DUT + availability, DUT capacity and packet loss in long term testing at + Throughput level, where some low-level of background loss may be present + and characterized. + + Tests have been (or will be) designed to collect the metrics + below: + + + Throughput Tests to measure the maximum forwarding rate (in + frames per second or fps) and bit rate (in Mbps) for a constant load + (as defined by ) without traffic loss. + + Packet and Frame Delay Distribution Tests to measure average, min + and max packet and frame delay for constant loads. + + Packet Delay Tests to understand latency distribution for + different packet sizes and over an extended test run to uncover + outliers. + + Scalability Tests to understand how the virtual switch performs + as the number of flows, active ports, complexity of the forwarding + logic’s configuration… it has to deal with + increases. + + Stream Performance Tests (TCP, UDP) to measure bulk data transfer + performance, i.e. how fast systems can send and receive data through + the switch. + + Control Path and Datapath Coupling Tests, to understand how + closely coupled the datapath and the control path are as well as the + effect of this coupling on the performance of the DUT (example: + delay of the initial packet of a flow). + + CPU and Memory Consumption Tests to understand the virtual + switch’s footprint on the system, usually conducted as + auxiliary measurements with benchmarks above. They include: CPU + utilization, Cache utilization and Memory footprint. + + The so-called "Soak" tests, where the selected test is conducted + over a long period of time (with an ideal duration of 24 hours, and + at least 6 hours). The purpose of soak tests is to capture transient + changes in performance which may occur due to infrequent processes + or the low probability coincidence of two or more processes. The + performance must be evaluated periodically during continuous + testing, and this results in use of Frame + Rate metrics instead of Throughput (which + requires stopping traffic to allow time for all traffic to exit + internal queues). + + + Future/planned test specs include: + Request/Response Performance Tests (TCP, UDP) which measure the + transaction rate through the switch. + + Noisy Neighbour Tests, to understand the effects of resource + sharing on the performance of a virtual switch. + + Tests derived from examination of ETSI NFV Draft GS IFA003 + requirements on characterization of + acceleration technologies applied to vswitches. + The flexibility of deployment of a virtual switch within a + network means that the BMWG IETF existing literature needs to be used to + characterize the performance of a switch in various deployment + scenarios. The deployment scenarios under consideration include: + +
+ Physical port to virtual switch to physical + port + + +
+ +
+ Physical port to virtual switch to VNF to virtual switch + to physical port + + +
+ Physical port to virtual switch to VNF to virtual switch + to VNF to virtual switch to physical port + + +
+ Physical port to virtual switch to VNF + + +
+ VNF to virtual switch to physical port + + +
+ VNF to virtual switch to VNF + + +
+ + A set of Deployment Scenario figures is available on the VSPERF Test + Methodology Wiki page . +
+ +
+ This section organizes the many existing test specifications into the + "3x3" matrix (introduced in ). + Because the LTD specification ID names are quite long, this section is + organized into lists for each occupied cell of the matrix (not all are + occupied, also the matrix has grown to 3x4 to accommodate scale metrics + when displaying the coverage of many metrics/benchmarks). The current + version of the LTD specification is available . + + The tests listed below assess the activation of paths in the data + plane, rather than the control plane. + + A complete list of tests with short summaries is available on the + VSPERF "LTD Test Spec Overview" Wiki page . + +
+ + Activation.RFC2889.AddressLearningRate + + PacketLatency.InitialPacketProcessingLatency + +
+ +
+ + CPDP.Coupling.Flow.Addition + +
+ +
+ + Throughput.RFC2544.SystemRecoveryTime + + Throughput.RFC2544.ResetTime + +
+ +
+ + Activation.RFC2889.AddressCachingCapacity + +
+ +
+ + Throughput.RFC2544.PacketLossRate + + CPU.RFC2544.0PacketLoss + + Throughput.RFC2544.PacketLossRateFrameModification + + Throughput.RFC2544.BackToBackFrames + + Throughput.RFC2889.MaxForwardingRate + + Throughput.RFC2889.ForwardPressure + + Throughput.RFC2889.BroadcastFrameForwarding + +
+ +
+ + Throughput.RFC2889.ErrorFramesFiltering + + Throughput.RFC2544.Profile + +
+ +
+ + Throughput.RFC2889.Soak + + Throughput.RFC2889.SoakFrameModification + + PacketDelayVariation.RFC3393.Soak + +
+ +
+ + Scalability.RFC2544.0PacketLoss + + MemoryBandwidth.RFC2544.0PacketLoss.Scalability + +
+ +
+
+ +
+
+
+ +
+ Benchmarking activities as described in this memo are limited to + technology characterization of a Device Under Test/System Under Test + (DUT/SUT) using controlled stimuli in a laboratory environment, with + dedicated address space and the constraints specified in the sections + above. + + The benchmarking network topology will be an independent test setup + and MUST NOT be connected to devices that may forward the test traffic + into a production network, or misroute traffic to the test management + network. + + Further, benchmarking is performed on a "black-box" basis, relying + solely on measurements observable external to the DUT/SUT. + + Special capabilities SHOULD NOT exist in the DUT/SUT specifically for + benchmarking purposes. Any implications for network security arising + from the DUT/SUT SHOULD be identical in the lab and in production + networks. +
+ +
+ No IANA Action is requested at this time. +
+ +
+ The authors appreciate and acknowledge comments from Scott Bradner, + Marius Georgescu, Ramki Krishnan, and Doug Montgomery, and others for + their reviews. +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Network Function Virtualization: Performance and Portability + Best Practices + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Test Topologies + https://wiki.opnfv.org/vsperf/test_methodology + + + + + + + + + + + + LTD Test Spec Overview + https://wiki.opnfv.org/wiki/vswitchperf_test_spec_review + + + + + + + + + + + + LTD Test Specification + http://artifacts.opnfv.org/vswitchperf/docs/requirements/index.html + + + + + + + + + + + + Brahmaputra, Second OPNFV Release + https://www.opnfv.org/brahmaputra + + + + + + + + + + + + https://docbox.etsi.org/ISG/NFV/Open/Drafts/IFA003_Acceleration_-_vSwitch_Spec/ + + + + + + + + + + + diff --git a/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-02.xml b/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-02.xml new file mode 100644 index 00000000..9157763e --- /dev/null +++ b/docs/testing/developer/requirements/ietf_draft/draft-vsperf-bmwg-vswitch-opnfv-02.xml @@ -0,0 +1,1016 @@ + + + + + + + + + + + + + + + Benchmarking Virtual Switches in + OPNFV + + + Intel + +
+ + + + + + + + + + + + + + + + + maryam.tahhan@intel.com + + +
+ + + + Intel + +
+ + + + + + + + + + + + + + + + + billy.o.mahony@intel.com + + +
+ + + + AT&T Labs + +
+ + 200 Laurel Avenue South + + Middletown, + + NJ + + 07748 + + USA + + + +1 732 420 1571 + + +1 732 368 1192 + + acmorton@att.com + + http://home.comcast.net/~acmacm/ +
+ + + + + + This memo describes the progress of the Open Platform for NFV (OPNFV) + project on virtual switch performance "VSWITCHPERF". This project + intends to build on the current and completed work of the Benchmarking + Methodology Working Group in IETF, by referencing existing literature. + The Benchmarking Methodology Working Group has traditionally conducted + laboratory characterization of dedicated physical implementations of + internetworking functions. Therefore, this memo begins to describe the + additional considerations when virtual switches are implemented in + general-purpose hardware. The expanded tests and benchmarks are also + influenced by the OPNFV mission to support virtualization of the "telco" + infrastructure. + + + + The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", + "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this + document are to be interpreted as described in RFC 2119. + + + + + + +
+ Benchmarking Methodology Working Group (BMWG) has traditionally + conducted laboratory characterization of dedicated physical + implementations of internetworking functions. The Black-box Benchmarks + of Throughput, Latency, Forwarding Rates and others have served our + industry for many years. Now, Network Function Virtualization (NFV) has + the goal to transform how internetwork functions are implemented, and + therefore has garnered much attention. + + This memo summarizes the progress of the Open Platform for NFV + (OPNFV) project on virtual switch performance characterization, + "VSWITCHPERF", through the Brahmaputra (second) release . This project intends to build on the current and + completed work of the Benchmarking Methodology Working Group in IETF, by + referencing existing literature. For example, currently the most often + referenced RFC is (which depends on ) and foundation of the benchmarking work in OPNFV is + common and strong. + + See + https://wiki.opnfv.org/characterize_vswitch_performance_for_telco_nfv_use_cases + for more background, and the OPNFV website for general information: + https://www.opnfv.org/ + + The authors note that OPNFV distinguishes itself from other open + source compute and networking projects through its emphasis on existing + "telco" services as opposed to cloud-computing. There are many ways in + which telco requirements have different emphasis on performance + dimensions when compared to cloud computing: support for and transfer of + isochronous media streams is one example. + + Note also that the move to NFV Infrastructure has resulted in many + new benchmarking initiatives across the industry. The authors are + currently doing their best to maintain alignment with many other + projects, and this Internet Draft is one part of the efforts. We + acknowledge the early work in , and useful + discussion with the authors. +
+ +
+ The primary purpose and scope of the memo is to inform the industry + of work-in-progress that builds on the body of extensive BMWG literature + and experience, and describe the extensions needed for benchmarking + virtual switches. Inital feedback indicates that many of these + extensions may be applicable beyond the current scope (to hardware + switches in the NFV Infrastructure and to virtual routers, for example). + Additionally, this memo serves as a vehicle to include more detail and + commentary from BMWG and other Open Source communities, under BMWG's + chartered work to characterize the NFV Infrastructure (a virtual switch + is an important aspect of that infrastructure). +
+ +
+ This section highlights some specific considerations (from )related to Benchmarks for virtual + switches. The OPNFV project is sharing its present view on these areas, + as they develop their specifications in the Level Test Design (LTD) + document. + +
+ To compare the performance of virtual designs and implementations + with their physical counterparts, identical benchmarks are needed. + BMWG has developed specifications for many network functions this memo + re-uses existing benchmarks through references, and expands them + during development of new methods. A key configuration aspect is the + number of parallel cores required to achieve comparable performance + with a given physical device, or whether some limit of scale was + reached before the cores could achieve the comparable level. + + It's unlikely that the virtual switch will be the only application + running on the SUT, so CPU utilization, Cache utilization, and Memory + footprint should also be recorded for the virtual implementations of + internetworking functions. +
+ +
+ External observations remain essential as the basis for Benchmarks. + Internal observations with fixed specification and interpretation will + be provided in parallel to assist the development of operations + procedures when the technology is deployed. +
+ +
+ A key consideration when conducting any sort of benchmark is trying + to ensure the consistency and repeatability of test results. When + benchmarking the performance of a vSwitch there are many factors that + can affect the consistency of results, one key factor is matching the + various hardware and software details of the SUT. This section lists + some of the many new parameters which this project believes are + critical to report in order to achieve repeatability. + + Hardware details including: + + + Platform details + + Processor details + + Memory information (type and size) + + Number of enabled cores + + Number of cores used for the test + + Number of physical NICs, as well as their details + (manufacturer, versions, type and the PCI slot they are plugged + into) + + NIC interrupt configuration + + BIOS version, release date and any configurations that were + modified + + CPU microcode level + + Memory DIMM configurations (quad rank performance may not be + the same as dual rank) in size, freq and slot locations + + PCI configuration parameters (payload size, early ack + option...) + + Power management at all levels (ACPI sleep states, processor + package, OS...) + Software details including: + + + OS parameters and behavior (text vs graphical no one typing at + the console on one system) + + OS version (for host and VNF) + + Kernel version (for host and VNF) + + GRUB boot parameters (for host and VNF) + + Hypervisor details (Type and version) + + Selected vSwitch, version number or commit id used + + vSwitch launch command line if it has been parameterised + + Memory allocation to the vSwitch + + which NUMA node it is using, and how many memory channels + + DPDK or any other SW dependency version number or commit id + used + + Memory allocation to a VM - if it's from Hugpages/elsewhere + + VM storage type: snapshot/independent persistent/independent + non-persistent + + Number of VMs + + Number of Virtual NICs (vNICs), versions, type and driver + + Number of virtual CPUs and their core affinity on the host + + Number vNIC interrupt configuration + + Thread affinitization for the applications (including the + vSwitch itself) on the host + + Details of Resource isolation, such as CPUs designated for + Host/Kernel (isolcpu) and CPUs designated for specific processes + (taskset). - Test duration. - Number of flows. + + + Test Traffic Information: + Traffic type - UDP, TCP, IMIX / Other + + Packet Sizes + + Deployment Scenario + + + +
+ +
+ Virtual switches group packets into flows by processing and + matching particular packet or frame header information, or by matching + packets based on the input ports. Thus a flow can be thought of a + sequence of packets that have the same set of header field values or + have arrived on the same port. Performance results can vary based on + the parameters the vSwitch uses to match for a flow. The recommended + flow classification parameters for any vSwitch performance tests are: + the input port, the source IP address, the destination IP address and + the Ethernet protocol type field. It is essential to increase the flow + timeout time on a vSwitch before conducting any performance tests that + do not measure the flow setup time. Normally the first packet of a + particular stream will install the flow in the virtual switch which + adds an additional latency, subsequent packets of the same flow are + not subject to this latency if the flow is already installed on the + vSwitch. +
+ +
+ This outline describes measurement of baseline with isolated + resources at a high level, which is the intended approach at this + time. + + + Baselines: + Optional: Benchmark platform forwarding capability without + a vswitch or VNF for at least 72 hours (serves as a means of + platform validation and a means to obtain the base performance + for the platform in terms of its maximum forwarding rate and + latency).
+ Benchmark platform forwarding + capability + + + + +
+ + Benchmark VNF forwarding capability with direct + connectivity (vSwitch bypass, e.g., SR/IOV) for at least 72 + hours (serves as a means of VNF validation and a means to + obtain the base performance for the VNF in terms of its + maximum forwarding rate and latency). The metrics gathered + from this test will serve as a key comparison point for + vSwitch bypass technologies performance and vSwitch + performance.
+ Benchmark VNF forwarding capability + + + + +
+ + Benchmarking with isolated resources alone, with other + resources (both HW&SW) disabled Example, vSw and VM are + SUT + + Benchmarking with isolated resources alone, leaving some + resources unused + + Benchmark with isolated resources and all resources + occupied + + + Next Steps + Limited sharing + + Production scenarios + + Stressful scenarios + + +
+
+ +
+ The overall specification in preparation is referred to as a Level + Test Design (LTD) document, which will contain a suite of performance + tests. The base performance tests in the LTD are based on the + pre-existing specifications developed by BMWG to test the performance of + physical switches. These specifications include: + + + Benchmarking Methodology for Network + Interconnect Devices + + Benchmarking Methodology for LAN + Switching + + Device Reset Characterization + + Packet Delay Variation Applicability + Statement + + + Some of the above/newer RFCs are being applied in benchmarking for + the first time, and represent a development challenge for test equipment + developers. Fortunately, many members of the testing system community + have engaged on the VSPERF project, including an open source test + system. + + In addition to this, the LTD also re-uses the terminology defined + by: + + + Benchmarking Terminology for LAN + Switching Devices + + Packet Delay Variation Applicability + Statement + + + + + Specifications to be included in future updates of the LTD + include: + Methodology for IP Multicast + Benchmarking + + Packet Reordering Metrics + + + As one might expect, the most fundamental internetworking + characteristics of Throughput and Latency remain important when the + switch is virtualized, and these benchmarks figure prominently in the + specification. + + When considering characteristics important to "telco" network + functions, we must begin to consider additional performance metrics. In + this case, the project specifications have referenced metrics from the + IETF IP Performance Metrics (IPPM) literature. This means that the test of Latency is replaced by measurement of a + metric derived from IPPM's , where a set of + statistical summaries will be provided (mean, max, min, etc.). Further + metrics planned to be benchmarked include packet delay variation as + defined by , reordering, burst behaviour, DUT + availability, DUT capacity and packet loss in long term testing at + Throughput level, where some low-level of background loss may be present + and characterized. + + Tests have been (or will be) designed to collect the metrics + below: + + + Throughput Tests to measure the maximum forwarding rate (in + frames per second or fps) and bit rate (in Mbps) for a constant load + (as defined by ) without traffic loss. + + Packet and Frame Delay Distribution Tests to measure average, min + and max packet and frame delay for constant loads. + + Packet Delay Tests to understand latency distribution for + different packet sizes and over an extended test run to uncover + outliers. + + Scalability Tests to understand how the virtual switch performs + as the number of flows, active ports, complexity of the forwarding + logic’s configuration… it has to deal with + increases. + + Stream Performance Tests (TCP, UDP) to measure bulk data transfer + performance, i.e. how fast systems can send and receive data through + the switch. + + Control Path and Datapath Coupling Tests, to understand how + closely coupled the datapath and the control path are as well as the + effect of this coupling on the performance of the DUT (example: + delay of the initial packet of a flow). + + CPU and Memory Consumption Tests to understand the virtual + switch’s footprint on the system, usually conducted as + auxiliary measurements with benchmarks above. They include: CPU + utilization, Cache utilization and Memory footprint. + + The so-called "Soak" tests, where the selected test is conducted + over a long period of time (with an ideal duration of 24 hours, and + at least 6 hours). The purpose of soak tests is to capture transient + changes in performance which may occur due to infrequent processes + or the low probability coincidence of two or more processes. The + performance must be evaluated periodically during continuous + testing, and this results in use of Frame + Rate metrics instead of Throughput (which + requires stopping traffic to allow time for all traffic to exit + internal queues). + + + Future/planned test specs include: + Request/Response Performance Tests (TCP, UDP) which measure the + transaction rate through the switch. + + Noisy Neighbour Tests, to understand the effects of resource + sharing on the performance of a virtual switch. + + Tests derived from examination of ETSI NFV Draft GS IFA003 + requirements on characterization of + acceleration technologies applied to vswitches. + The flexibility of deployment of a virtual switch within a + network means that the BMWG IETF existing literature needs to be used to + characterize the performance of a switch in various deployment + scenarios. The deployment scenarios under consideration include: + +
+ Physical port to virtual switch to physical + port + + +
+ +
+ Physical port to virtual switch to VNF to virtual switch + to physical port + + +
+ Physical port to virtual switch to VNF to virtual switch + to VNF to virtual switch to physical port + + +
+ Physical port to virtual switch to VNF + + +
+ VNF to virtual switch to physical port + + +
+ VNF to virtual switch to VNF + + +
+ + A set of Deployment Scenario figures is available on the VSPERF Test + Methodology Wiki page . +
+ +
+ This section organizes the many existing test specifications into the + "3x3" matrix (introduced in ). + Because the LTD specification ID names are quite long, this section is + organized into lists for each occupied cell of the matrix (not all are + occupied, also the matrix has grown to 3x4 to accommodate scale metrics + when displaying the coverage of many metrics/benchmarks). The current + version of the LTD specification is available . + + The tests listed below assess the activation of paths in the data + plane, rather than the control plane. + + A complete list of tests with short summaries is available on the + VSPERF "LTD Test Spec Overview" Wiki page . + +
+ + Activation.RFC2889.AddressLearningRate + + PacketLatency.InitialPacketProcessingLatency + +
+ +
+ + CPDP.Coupling.Flow.Addition + +
+ +
+ + Throughput.RFC2544.SystemRecoveryTime + + Throughput.RFC2544.ResetTime + +
+ +
+ + Activation.RFC2889.AddressCachingCapacity + +
+ +
+ + Throughput.RFC2544.PacketLossRate + + CPU.RFC2544.0PacketLoss + + Throughput.RFC2544.PacketLossRateFrameModification + + Throughput.RFC2544.BackToBackFrames + + Throughput.RFC2889.MaxForwardingRate + + Throughput.RFC2889.ForwardPressure + + Throughput.RFC2889.BroadcastFrameForwarding + +
+ +
+ + Throughput.RFC2889.ErrorFramesFiltering + + Throughput.RFC2544.Profile + +
+ +
+ + Throughput.RFC2889.Soak + + Throughput.RFC2889.SoakFrameModification + + PacketDelayVariation.RFC3393.Soak + +
+ +
+ + Scalability.RFC2544.0PacketLoss + + MemoryBandwidth.RFC2544.0PacketLoss.Scalability + +
+ +
+
+ +
+
+
+ +
+ Benchmarking activities as described in this memo are limited to + technology characterization of a Device Under Test/System Under Test + (DUT/SUT) using controlled stimuli in a laboratory environment, with + dedicated address space and the constraints specified in the sections + above. + + The benchmarking network topology will be an independent test setup + and MUST NOT be connected to devices that may forward the test traffic + into a production network, or misroute traffic to the test management + network. + + Further, benchmarking is performed on a "black-box" basis, relying + solely on measurements observable external to the DUT/SUT. + + Special capabilities SHOULD NOT exist in the DUT/SUT specifically for + benchmarking purposes. Any implications for network security arising + from the DUT/SUT SHOULD be identical in the lab and in production + networks. +
+ +
+ No IANA Action is requested at this time. +
+ +
+ The authors appreciate and acknowledge comments from Scott Bradner, + Marius Georgescu, Ramki Krishnan, Doug Montgomery, Martin Klozik, + Christian Trautman, and others for their reviews. +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Network Function Virtualization: Performance and Portability + Best Practices + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Test Topologies + https://wiki.opnfv.org/vsperf/test_methodology + + + + + + + + + + + + LTD Test Spec Overview + https://wiki.opnfv.org/wiki/vswitchperf_test_spec_review + + + + + + + + + + + + LTD Test Specification + http://artifacts.opnfv.org/vswitchperf/docs/requirements/index.html + + + + + + + + + + + + Brahmaputra, Second OPNFV Release + https://www.opnfv.org/brahmaputra + + + + + + + + + + + + https://docbox.etsi.org/ISG/NFV/Open/Drafts/IFA003_Acceleration_-_vSwitch_Spec/ + + + + + + + + + + + diff --git a/docs/testing/developer/requirements/vm2vm_alternative_benchmark.png b/docs/testing/developer/requirements/vm2vm_alternative_benchmark.png new file mode 100644 index 00000000..d21334ba Binary files /dev/null and b/docs/testing/developer/requirements/vm2vm_alternative_benchmark.png differ diff --git a/docs/testing/developer/requirements/vm2vm_benchmark.png b/docs/testing/developer/requirements/vm2vm_benchmark.png new file mode 100644 index 00000000..3a85e51f Binary files /dev/null and b/docs/testing/developer/requirements/vm2vm_benchmark.png differ diff --git a/docs/testing/developer/requirements/vm2vm_hypervisor_benchmark.png b/docs/testing/developer/requirements/vm2vm_hypervisor_benchmark.png new file mode 100644 index 00000000..b5b76e8a Binary files /dev/null and b/docs/testing/developer/requirements/vm2vm_hypervisor_benchmark.png differ diff --git a/docs/testing/developer/requirements/vm2vm_virtual_interface_benchmark.png b/docs/testing/developer/requirements/vm2vm_virtual_interface_benchmark.png new file mode 100644 index 00000000..55294af6 Binary files /dev/null and b/docs/testing/developer/requirements/vm2vm_virtual_interface_benchmark.png differ diff --git a/docs/testing/developer/requirements/vswitchperf_ltd.rst b/docs/testing/developer/requirements/vswitchperf_ltd.rst new file mode 100644 index 00000000..e1372520 --- /dev/null +++ b/docs/testing/developer/requirements/vswitchperf_ltd.rst @@ -0,0 +1,1712 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intel Corporation, AT&T and others. + +****************************** +VSPERF LEVEL TEST DESIGN (LTD) +****************************** + +.. 3.1 + +============ +Introduction +============ + +The intention of this Level Test Design (LTD) document is to specify the set of +tests to carry out in order to objectively measure the current characteristics +of a virtual switch in the Network Function Virtualization Infrastructure +(NFVI) as well as the test pass criteria. The detailed test cases will be +defined in details-of-LTD_, preceded by the doc-id-of-LTD_ and the scope-of-LTD_. + +This document is currently in draft form. + +.. 3.1.1 + + +.. _doc-id-of-LTD: + +Document identifier +=================== + +The document id will be used to uniquely +identify versions of the LTD. The format for the document id will be: +OPNFV\_vswitchperf\_LTD\_REL\_STATUS, where by the +status is one of: draft, reviewed, corrected or final. The document id +for this version of the LTD is: +OPNFV\_vswitchperf\_LTD\_Brahmaputra\_REVIEWED. + +.. 3.1.2 + +.. _scope-of-LTD: + +Scope +===== + +The main purpose of this project is to specify a suite of +performance tests in order to objectively measure the current packet +transfer characteristics of a virtual switch in the NFVI. The intent of +the project is to facilitate testing of any virtual switch. Thus, a +generic suite of tests shall be developed, with no hard dependencies to +a single implementation. In addition, the test case suite shall be +architecture independent. + +The test cases developed in this project shall not form part of a +separate test framework, all of these tests may be inserted into the +Continuous Integration Test Framework and/or the Platform Functionality +Test Framework - if a vSwitch becomes a standard component of an OPNFV +release. + +.. 3.1.3 + +References +========== + +* `RFC 1242 Benchmarking Terminology for Network Interconnection + Devices `__ +* `RFC 2544 Benchmarking Methodology for Network Interconnect + Devices `__ +* `RFC 2285 Benchmarking Terminology for LAN Switching + Devices `__ +* `RFC 2889 Benchmarking Methodology for LAN Switching + Devices `__ +* `RFC 3918 Methodology for IP Multicast + Benchmarking `__ +* `RFC 4737 Packet Reordering + Metrics `__ +* `RFC 5481 Packet Delay Variation Applicability + Statement `__ +* `RFC 6201 Device Reset + Characterization `__ + +.. 3.2 + +.. _details-of-LTD: + +================================ +Details of the Level Test Design +================================ + +This section describes the features to be tested (FeaturesToBeTested-of-LTD_), and +identifies the sets of test cases or scenarios (TestIdentification-of-LTD_). + +.. 3.2.1 + +.. _FeaturesToBeTested-of-LTD: + +Features to be tested +===================== + +Characterizing virtual switches (i.e. Device Under Test (DUT) in this document) +includes measuring the following performance metrics: + +- Throughput +- Packet delay +- Packet delay variation +- Packet loss +- Burst behaviour +- Packet re-ordering +- Packet correctness +- Availability and capacity of the DUT + +.. 3.2.2 + +.. _TestIdentification-of-LTD: + +Test identification +=================== + +.. 3.2.2.1 + +Throughput tests +---------------- + +The following tests aim to determine the maximum forwarding rate that +can be achieved with a virtual switch. The list is not exhaustive but +should indicate the type of tests that should be required. It is +expected that more will be added. + +.. 3.2.2.1.1 + +.. _PacketLossRatio: + +Test ID: LTD.Throughput.RFC2544.PacketLossRatio +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 X% packet loss ratio Throughput and Latency Test + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + This test determines the DUT's maximum forwarding rate with X% traffic + loss for a constant load (fixed length frames at a fixed interval time). + The default loss percentages to be tested are: - X = 0% - X = 10^-7% + + Note: Other values can be tested if required by the user. + + The selected frame sizes are those previously defined under + :ref:`default-test-parameters`. + The test can also be used to determine the average latency of the traffic. + + Under the `RFC2544 `__ + test methodology, the test duration will + include a number of trials; each trial should run for a minimum period + of 60 seconds. A binary search methodology must be applied for each + trial to obtain the final result. + + **Expected Result**: At the end of each trial, the presence or absence + of loss determines the modification of offered load for the next trial, + converging on a maximum rate, or + `RFC2544 `__ Throughput with X% + loss. + The Throughput load is re-used in related + `RFC2544 `__ tests and other + tests. + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of + the DUT for each frame size with X% packet loss. + - The average latency of the traffic flow when passing through the DUT + (if testing for latency, note that this average is different from the + test specified in Section 26.3 of + `RFC2544 `__). + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + +.. 3.2.2.1.2 + +.. _PacketLossRatioFrameModification: + +Test ID: LTD.Throughput.RFC2544.PacketLossRatioFrameModification +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 X% packet loss Throughput and Latency Test with + packet modification + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + This test determines the DUT's maximum forwarding rate with X% traffic + loss for a constant load (fixed length frames at a fixed interval time). + The default loss percentages to be tested are: - X = 0% - X = 10^-7% + + Note: Other values can be tested if required by the user. + + The selected frame sizes are those previously defined under + :ref:`default-test-parameters`. + The test can also be used to determine the average latency of the traffic. + + Under the `RFC2544 `__ + test methodology, the test duration will + include a number of trials; each trial should run for a minimum period + of 60 seconds. A binary search methodology must be applied for each + trial to obtain the final result. + + During this test, the DUT must perform the following operations on the + traffic flow: + + - Perform packet parsing on the DUT's ingress port. + - Perform any relevant address look-ups on the DUT's ingress ports. + - Modify the packet header before forwarding the packet to the DUT's + egress port. Packet modifications include: + + - Modifying the Ethernet source or destination MAC address. + - Modifying/adding a VLAN tag. (**Recommended**). + - Modifying/adding a MPLS tag. + - Modifying the source or destination ip address. + - Modifying the TOS/DSCP field. + - Modifying the source or destination ports for UDP/TCP/SCTP. + - Modifying the TTL. + + **Expected Result**: The Packet parsing/modifications require some + additional degree of processing resource, therefore the + `RFC2544 `__ + Throughput is expected to be somewhat lower than the Throughput level + measured without additional steps. The reduction is expected to be + greatest on tests with the smallest packet sizes (greatest header + processing rates). + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of + the DUT for each frame size with X% packet loss and packet + modification operations being performed by the DUT. + - The average latency of the traffic flow when passing through the DUT + (if testing for latency, note that this average is different from the + test specified in Section 26.3 of + `RFC2544 `__). + - The `RFC5481 `__ + PDV form of delay variation on the traffic flow, + using the 99th percentile. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + +.. 3.2.2.1.3 + +Test ID: LTD.Throughput.RFC2544.Profile +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 Throughput and Latency Profile + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + This test reveals how throughput and latency degrades as the offered + rate varies in the region of the DUT's maximum forwarding rate as + determined by LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss). + For example it can be used to determine if the degradation of throughput + and latency as the offered rate increases is slow and graceful or sudden + and severe. + + The selected frame sizes are those previously defined under + :ref:`default-test-parameters`. + + The offered traffic rate is described as a percentage delta with respect + to the DUT's RFC 2544 Throughput as determined by + LTD.Throughput.RFC2544.PacketLoss Ratio (0% Packet Loss case). A delta + of 0% is equivalent to an offered traffic rate equal to the RFC 2544 + Maximum Throughput; A delta of +50% indicates an offered rate half-way + between the Maximum RFC2544 Throughput and line-rate, whereas a delta of + -50% indicates an offered rate of half the RFC 2544 Maximum Throughput. + Therefore the range of the delta figure is natuarlly bounded at -100% + (zero offered traffic) and +100% (traffic offered at line rate). + + The following deltas to the maximum forwarding rate should be applied: + + - -50%, -10%, 0%, +10% & +50% + + **Expected Result**: For each packet size a profile should be produced + of how throughput and latency vary with offered rate. + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The forwarding rate in Frames Per Second (FPS) and Mbps of the DUT + for each delta to the maximum forwarding rate and for each frame + size. + - The average latency for each delta to the maximum forwarding rate and + for each frame size. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + - Any failures experienced (for example if the vSwitch crashes, stops + processing packets, restarts or becomes unresponsive to commands) + when the offered load is above Maximum Throughput MUST be recorded + and reported with the results. + +.. 3.2.2.1.4 + +Test ID: LTD.Throughput.RFC2544.SystemRecoveryTime +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 System Recovery Time Test + + **Prerequisite Test** LTD.Throughput.RFC2544.PacketLossRatio + + **Priority**: + + **Description**: + + The aim of this test is to determine the length of time it takes the DUT + to recover from an overload condition for a constant load (fixed length + frames at a fixed interval time). The selected frame sizes are those + previously defined under :ref:`default-test-parameters`, + traffic should be sent to the DUT under normal conditions. During the + duration of the test and while the traffic flows are passing though the + DUT, at least one situation leading to an overload condition for the DUT + should occur. The time from the end of the overload condition to when + the DUT returns to normal operations should be measured to determine + recovery time. Prior to overloading the DUT, one should record the + average latency for 10,000 packets forwarded through the DUT. + + The overload condition SHOULD be to transmit traffic at a very high + frame rate to the DUT (150% of the maximum 0% packet loss rate as + determined by LTD.Throughput.RFC2544.PacketLossRatio or line-rate + whichever is lower), for at least 60 seconds, then reduce the frame rate + to 75% of the maximum 0% packet loss rate. A number of time-stamps + should be recorded: - Record the time-stamp at which the frame rate was + reduced and record a second time-stamp at the time of the last frame + lost. The recovery time is the difference between the two timestamps. - + Record the average latency for 10,000 frames after the last frame loss + and continue to record average latency measurements for every 10,000 + frames, when latency returns to within 10% of pre-overload levels record + the time-stamp. + + **Expected Result**: + + **Metrics collected** + + The following are the metrics collected for this test: + + - The length of time it takes the DUT to recover from an overload + condition. + - The length of time it takes the DUT to recover the average latency to + pre-overload conditions. + + **Deployment scenario**: + + - Physical → virtual switch → physical. + +.. 3.2.2.1.5 + +.. _BackToBackFrames: + +Test ID: LTD.Throughput.RFC2544.BackToBackFrames +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2544 Back To Back Frames Test + + **Prerequisite Test**: N + + **Priority**: + + **Description**: + + The aim of this test is to characterize the ability of the DUT to + process back-to-back frames. For each frame size previously defined + under :ref:`default-test-parameters`, a burst of traffic + is sent to the DUT with the minimum inter-frame gap between each frame. + If the number of received frames equals the number of frames that were + transmitted, the burst size should be increased and traffic is sent to + the DUT again. The value measured is the back-to-back value, that is the + maximum burst size the DUT can handle without any frame loss. Please note + a trial must run for a minimum of 2 seconds and should be repeated 50 + times (at a minimum). + + **Expected Result**: + + Tests of back-to-back frames with physical devices have produced + unstable results in some cases. All tests should be repeated in multiple + test sessions and results stability should be examined. + + **Metrics collected** + + The following are the metrics collected for this test: + + - The average back-to-back value across the trials, which is + the number of frames in the longest burst that the DUT will + handle without the loss of any frames. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + + **Deployment scenario**: + + - Physical → virtual switch → physical. + +.. 3.2.2.1.6 + +Test ID: LTD.Throughput.RFC2889.MaxForwardingRateSoak +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2889 X% packet loss Max Forwarding Rate Soak Test + + **Prerequisite Test** LTD.Throughput.RFC2544.PacketLossRatio + + **Priority**: + + **Description**: + + The aim of this test is to understand the Max Forwarding Rate stability + over an extended test duration in order to uncover any outliers. To allow + for an extended test duration, the test should ideally run for 24 hours + or, if this is not possible, for at least 6 hours. For this test, each frame + size must be sent at the highest Throughput rate with X% packet loss, as + determined in the prerequisite test. The default loss percentages to be + tested are: - X = 0% - X = 10^-7% + + Note: Other values can be tested if required by the user. + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - Max Forwarding Rate stability of the DUT. + + - This means reporting the number of packets lost per time interval + and reporting any time intervals with packet loss. The + `RFC2889 `__ + Forwarding Rate shall be measured in each interval. + An interval of 60s is suggested. + + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + - The `RFC5481 `__ + PDV form of delay variation on the traffic flow, + using the 99th percentile. + +.. 3.2.2.1.7 + +Test ID: LTD.Throughput.RFC2889.MaxForwardingRateSoakFrameModification +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2889 Max Forwarding Rate Soak Test with Frame Modification + + **Prerequisite Test**: + LTD.Throughput.RFC2544.PacketLossRatioFrameModification (0% Packet Loss) + + **Priority**: + + **Description**: + + The aim of this test is to understand the Max Forwarding Rate stability over an + extended test duration in order to uncover any outliers. To allow for an + extended test duration, the test should ideally run for 24 hours or, if + this is not possible, for at least 6 hour. For this test, each frame + size must be sent at the highest Throughput rate with 0% packet loss, as + determined in the prerequisite test. + + During this test, the DUT must perform the following operations on the + traffic flow: + + - Perform packet parsing on the DUT's ingress port. + - Perform any relevant address look-ups on the DUT's ingress ports. + - Modify the packet header before forwarding the packet to the DUT's + egress port. Packet modifications include: + + - Modifying the Ethernet source or destination MAC address. + - Modifying/adding a VLAN tag (**Recommended**). + - Modifying/adding a MPLS tag. + - Modifying the source or destination ip address. + - Modifying the TOS/DSCP field. + - Modifying the source or destination ports for UDP/TCP/SCTP. + - Modifying the TTL. + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - Max Forwarding Rate stability of the DUT. + + - This means reporting the number of packets lost per time interval + and reporting any time intervals with packet loss. The + `RFC2889 `__ + Forwarding Rate shall be measured in each interval. + An interval of 60s is suggested. + + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + - The `RFC5481 `__ + PDV form of delay variation on the traffic flow, using the 99th + percentile. + +.. 3.2.2.1.8 + +Test ID: LTD.Throughput.RFC6201.ResetTime +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 6201 Reset Time Test + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + The aim of this test is to determine the length of time it takes the DUT + to recover from a reset. + + Two reset methods are defined - planned and unplanned. A planned reset + requires stopping and restarting the virtual switch by the usual + 'graceful' method defined by it's documentation. An unplanned reset + requires simulating a fatal internal fault in the virtual switch - for + example by using kill -SIGKILL on a Linux environment. + + Both reset methods SHOULD be exercised. + + For each frame size previously defined under :ref:`default-test-parameters`, + traffic should be sent to the DUT under + normal conditions. During the duration of the test and while the traffic + flows are passing through the DUT, the DUT should be reset and the Reset + time measured. The Reset time is the total time that a device is + determined to be out of operation and includes the time to perform the + reset and the time to recover from it (cf. `RFC6201 + `__). + + `RFC6201 `__ defines two methods + to measure the Reset time: + + - Frame-Loss Method: which requires the monitoring of the number of + lost frames and calculates the Reset time based on the number of + frames lost and the offered rate according to the following + formula: + + .. code-block:: console + + Frames_lost (packets) + Reset_time = ------------------------------------- + Offered_rate (packets per second) + + - Timestamp Method: which measures the time from which the last frame + is forwarded from the DUT to the time the first frame is forwarded + after the reset. This involves time-stamping all transmitted frames + and recording the timestamp of the last frame that was received prior + to the reset and also measuring the timestamp of the first frame that + is received after the reset. The Reset time is the difference between + these two timestamps. + + According to `RFC6201 `__ the + choice of method depends on the test tool's capability; the Frame-Loss + method SHOULD be used if the test tool supports: + + * Counting the number of lost frames per stream. + * Transmitting test frame despite the physical link status. + + whereas the Timestamp method SHOULD be used if the test tool supports: + + * Timestamping each frame. + * Monitoring received frame's timestamp. + * Transmitting frames only if the physical link status is up. + + **Expected Result**: + + **Metrics collected** + + The following are the metrics collected for this test: + + * Average Reset Time over the number of trials performed. + + Results of this test should include the following information: + + * The reset method used. + * Throughput in Fps and Mbps. + * Average Frame Loss over the number of trials performed. + * Average Reset Time in milliseconds over the number of trials performed. + * Number of trials performed. + * Protocol: IPv4, IPv6, MPLS, etc. + * Frame Size in Octets + * Port Media: Ethernet, Gigabit Ethernet (GbE), etc. + * Port Speed: 10 Gbps, 40 Gbps etc. + * Interface Encapsulation: Ethernet, Ethernet VLAN, etc. + + **Deployment scenario**: + + * Physical → virtual switch → physical. + +.. 3.2.2.1.9 + +Test ID: LTD.Throughput.RFC2889.MaxForwardingRate +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2889 Forwarding Rate Test + + **Prerequisite Test**: LTD.Throughput.RFC2544.PacketLossRatio + + **Priority**: + + **Description**: + + This test measures the DUT's Max Forwarding Rate when the Offered Load + is varied between the throughput and the Maximum Offered Load for fixed + length frames at a fixed time interval. The selected frame sizes are + those previously defined under :ref:`default-test-parameters`. + The throughput is the maximum offered + load with 0% frame loss (measured by the prerequisite test), and the + Maximum Offered Load (as defined by + `RFC2285 `__) is *"the highest + number of frames per second that an external source can transmit to a + DUT/SUT for forwarding to a specified output interface or interfaces"*. + + Traffic should be sent to the DUT at a particular rate (TX rate) + starting with TX rate equal to the throughput rate. The rate of + successfully received frames at the destination counted (in FPS). If the + RX rate is equal to the TX rate, the TX rate should be increased by a + fixed step size and the RX rate measured again until the Max Forwarding + Rate is found. + + The trial duration for each iteration should last for the period of time + needed for the system to reach steady state for the frame size being + tested. Under `RFC2889 `__ + (Sec. 5.6.3.1) test methodology, the test + duration should run for a minimum period of 30 seconds, regardless + whether the system reaches steady state before the minimum duration + ends. + + **Expected Result**: According to + `RFC2889 `__ The Max Forwarding + Rate is the highest forwarding rate of a DUT taken from an iterative set of + forwarding rate measurements. The iterative set of forwarding rate measurements + are made by setting the intended load transmitted from an external source and + measuring the offered load (i.e what the DUT is capable of forwarding). If the + Throughput == the Maximum Offered Load, it follows that Max Forwarding Rate is + equal to the Maximum Offered Load. + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The Max Forwarding Rate for the DUT for each packet size. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + + **Deployment scenario**: + + - Physical → virtual switch → physical. Note: Full mesh tests with + multiple ingress and egress ports are a key aspect of RFC 2889 + benchmarks, and scenarios with both 2 and 4 ports should be tested. + In any case, the number of ports used must be reported. + +.. 3.2.2.1.10 + +Test ID: LTD.Throughput.RFC2889.ForwardPressure +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2889 Forward Pressure Test + + **Prerequisite Test**: LTD.Throughput.RFC2889.MaxForwardingRate + + **Priority**: + + **Description**: + + The aim of this test is to determine if the DUT transmits frames with an + inter-frame gap that is less than 12 bytes. This test overloads the DUT + and measures the output for forward pressure. Traffic should be + transmitted to the DUT with an inter-frame gap of 11 bytes, this will + overload the DUT by 1 byte per frame. The forwarding rate of the DUT + should be measured. + + **Expected Result**: The forwarding rate should not exceed the maximum + forwarding rate of the DUT collected by + LTD.Throughput.RFC2889.MaxForwardingRate. + + **Metrics collected** + + The following are the metrics collected for this test: + + - Forwarding rate of the DUT in FPS or Mbps. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + + **Deployment scenario**: + + - Physical → virtual switch → physical. + +.. 3.2.2.1.11 + +Test ID: LTD.Throughput.RFC2889.ErrorFramesFiltering +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2889 Error Frames Filtering Test + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + The aim of this test is to determine whether the DUT will propagate any + erroneous frames it receives or whether it is capable of filtering out + the erroneous frames. Traffic should be sent with erroneous frames + included within the flow at random intervals. Illegal frames that must + be tested include: - Oversize Frames. - Undersize Frames. - CRC Errored + Frames. - Dribble Bit Errored Frames - Alignment Errored Frames + + The traffic flow exiting the DUT should be recorded and checked to + determine if the erroneous frames where passed through the DUT. + + **Expected Result**: Broken frames are not passed! + + **Metrics collected** + + No Metrics are collected in this test, instead it determines: + + - Whether the DUT will propagate erroneous frames. + - Or whether the DUT will correctly filter out any erroneous frames + from traffic flow with out removing correct frames. + + **Deployment scenario**: + + - Physical → virtual switch → physical. + +.. 3.2.2.1.12 + +Test ID: LTD.Throughput.RFC2889.BroadcastFrameForwarding +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2889 Broadcast Frame Forwarding Test + + **Prerequisite Test**: N + + **Priority**: + + **Description**: + + The aim of this test is to determine the maximum forwarding rate of the + DUT when forwarding broadcast traffic. For each frame previously defined + under :ref:`default-test-parameters`, the traffic should + be set up as broadcast traffic. The traffic throughput of the DUT should + be measured. + + The test should be conducted with at least 4 physical ports on the DUT. + The number of ports used MUST be recorded. + + As broadcast involves forwarding a single incoming packet to several + destinations, the latency of a single packet is defined as the average + of the latencies for each of the broadcast destinations. + + The incoming packet is transmitted on each of the other physical ports, + it is not transmitted on the port on which it was received. The test MAY + be conducted using different broadcasting ports to uncover any + performance differences. + + **Expected Result**: + + **Metrics collected**: + + The following are the metrics collected for this test: + + - The forwarding rate of the DUT when forwarding broadcast traffic. + - The minimum, average & maximum packets latencies observed. + + **Deployment scenario**: + + - Physical → virtual switch 3x physical. In the Broadcast rate testing, + four test ports are required. One of the ports is connected to the test + device, so it can send broadcast frames and listen for miss-routed frames. + +.. 3.2.2.1.13 + +Test ID: LTD.Throughput.RFC2544.WorstN-BestN +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: Modified RFC 2544 X% packet loss ratio Throughput and Latency Test + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + This test determines the DUT's maximum forwarding rate with X% traffic + loss for a constant load (fixed length frames at a fixed interval time). + The default loss percentages to be tested are: X = 0%, X = 10^-7% + + Modified RFC 2544 throughput benchmarking methodology aims to quantify + the throughput measurement variations observed during standard RFC 2544 + benchmarking measurements of virtual switches and VNFs. The RFC2544 + binary search algorithm is modified to use more samples per test trial + to drive the binary search and yield statistically more meaningful + results. This keeps the heart of the RFC2544 methodology, still relying + on the binary search of throughput at specified loss tolerance, while + providing more useful information about the range of results seen in + testing. Instead of using a single traffic trial per iteration step, + each traffic trial is repeated N times and the success/failure of the + iteration step is based on these N traffic trials. Two types of revised + tests are defined - *Worst-of-N* and *Best-of-N*. + + **Worst-of-N** + + *Worst-of-N* indicates the lowest expected maximum throughput for ( + packet size, loss tolerance) when repeating the test. + + 1. Repeat the same test run N times at a set packet rate, record each + result. + 2. Take the WORST result (highest packet loss) out of N result samples, + called the Worst-of-N sample. + 3. If Worst-of-N sample has loss less than the set loss tolerance, then + the step is successful - increase the test traffic rate. + 4. If Worst-of-N sample has loss greater than the set loss tolerance + then the step failed - decrease the test traffic rate. + 5. Go to step 1. + + **Best-of-N** + + *Best-of-N* indicates the highest expected maximum throughput for ( + packet size, loss tolerance) when repeating the test. + + 1. Repeat the same traffic run N times at a set packet rate, record + each result. + 2. Take the BEST result (least packet loss) out of N result samples, + called the Best-of-N sample. + 3. If Best-of-N sample has loss less than the set loss tolerance, then + the step is successful - increase the test traffic rate. + 4. If Best-of-N sample has loss greater than the set loss tolerance, + then the step failed - decrease the test traffic rate. + 5. Go to step 1. + + Performing both Worst-of-N and Best-of-N benchmark tests yields lower + and upper bounds of expected maximum throughput under the operating + conditions, giving a very good indication to the user of the + deterministic performance range for the tested setup. + + **Expected Result**: At the end of each trial series, the presence or + absence of loss determines the modification of offered load for the + next trial series, converging on a maximum rate, or + `RFC2544 `__ Throughput + with X% loss. + The Throughput load is re-used in related + `RFC2544 `__ tests and other + tests. + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of + the DUT for each frame size with X% packet loss. + - The average latency of the traffic flow when passing through the DUT + (if testing for latency, note that this average is different from the + test specified in Section 26.3 of + `RFC2544 `__). + - Following may also be collected as part of this test, to determine + the vSwitch's performance footprint on the system: + + - CPU core utilization. + - CPU cache utilization. + - Memory footprint. + - System bus (QPI, PCI, ...) utilization. + - CPU cycles consumed per packet. + +.. 3.2.2.1.14 + +Test ID: LTD.Throughput.Overlay.Network..RFC2544.PacketLossRatio +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: Overlay Network RFC 2544 X% packet loss ratio Throughput and Latency Test + + + NOTE: Throughout this test, four interchangeable overlay technologies are covered by the + same test description. They are: VXLAN, GRE, NVGRE and GENEVE. + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + This test evaluates standard switch performance benchmarks for the scenario where an + Overlay Network is deployed for all paths through the vSwitch. Overlay Technologies covered + (replacing in the test name) include: + + - VXLAN + - GRE + - NVGRE + - GENEVE + + Performance will be assessed for each of the following overlay network functions: + + - Encapsulation only + - De-encapsulation only + - Both Encapsulation and De-encapsulation + + For each native packet, the DUT must perform the following operations: + + - Examine the packet and classify its correct overlay net (tunnel) assignment + - Encapsulate the packet + - Switch the packet to the correct port + + For each encapsulated packet, the DUT must perform the following operations: + + - Examine the packet and classify its correct native network assignment + - De-encapsulate the packet, if required + - Switch the packet to the correct port + + The selected frame sizes are those previously defined under + :ref:`default-test-parameters`. + + Thus, each test comprises an overlay technology, a network function, + and a packet size *with* overlay network overhead included + (but see also the discussion at + https://etherpad.opnfv.org/p/vSwitchTestsDrafts ). + + The test can also be used to determine the average latency of the traffic. + + Under the `RFC2544 `__ + test methodology, the test duration will + include a number of trials; each trial should run for a minimum period + of 60 seconds. A binary search methodology must be applied for each + trial to obtain the final result for Throughput. + + **Expected Result**: At the end of each trial, the presence or absence + of loss determines the modification of offered load for the next trial, + converging on a maximum rate, or + `RFC2544 `__ Throughput with X% + loss (where the value of X is typically equal to zero). + The Throughput load is re-used in related + `RFC2544 `__ tests and other + tests. + + **Metrics Collected**: + The following are the metrics collected for this test: + + - The maximum Throughput in Frames Per Second (FPS) and Mbps of + the DUT for each frame size with X% packet loss. + - The average latency of the traffic flow when passing through the DUT + and VNFs (if testing for latency, note that this average is different from the + test specified in Section 26.3 of + `RFC2544 `__). + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + +.. 3.2.3.1.15 + +Test ID: LTD.Throughput.RFC2544.MatchAction.PacketLossRatio +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 X% packet loss ratio match action Throughput and Latency Test + + **Prerequisite Test**: LTD.Throughput.RFC2544.PacketLossRatio + + **Priority**: + + **Description**: + + The aim of this test is to determine the cost of carrying out match + action(s) on the DUT’s RFC2544 Throughput with X% traffic loss for + a constant load (fixed length frames at a fixed interval time). + + Each test case requires: + + * selection of a specific match action(s), + * specifying a percentage of total traffic that is elligible + for the match action, + * determination of the specific test configuration (number + of flows, number of test ports, presence of an external + controller, etc.), and + * measurement of the RFC 2544 Throughput level with X% packet + loss: Traffic shall be bi-directional and symmetric. + + Note: It would be ideal to verify that all match action-elligible + traffic was forwarded to the correct port, and if forwarded to + an unintended port it should be considered lost. + + A match action is an action that is typically carried on a frame + or packet that matches a set of flow classification parameters + (typically frame/packet header fields). A match action may or may + not modify a packet/frame. Match actions include [1]: + + * output : outputs a packet to a particular port. + * normal: Subjects the packet to traditional L2/L3 processing + (MAC learning). + * flood: Outputs the packet on all switch physical ports + other than the port on which it was received and any ports + on which flooding is disabled. + * all: Outputs the packet on all switch physical ports other + than the port on which it was received. + * local: Outputs the packet on the ``local port``, which + corresponds to the network device that has the same name as + the bridge. + * in_port: Outputs the packet on the port from which it was + received. + * Controller: Sends the packet and its metadata to the + OpenFlow controller as a ``packet in`` message. + * enqueue: Enqueues the packet on the specified queue + within port. + * drop: discard the packet. + + Modifications include [1]: + + * mod vlan: covered by LTD.Throughput.RFC2544.PacketLossRatioFrameModification + * mod_dl_src: Sets the source Ethernet address. + * mod_dl_dst: Sets the destination Ethernet address. + * mod_nw_src: Sets the IPv4 source address. + * mod_nw_dst: Sets the IPv4 destination address. + * mod_tp_src: Sets the TCP or UDP or SCTP source port. + * mod_tp_dst: Sets the TCP or UDP or SCTP destination port. + * mod_nw_tos: Sets the DSCP bits in the IPv4 ToS/DSCP or + IPv6 traffic class field. + * mod_nw_ecn: Sets the ECN bits in the appropriate IPv4 or + IPv6 field. + * mod_nw_ttl: Sets the IPv4 TTL or IPv6 hop limit field. + + Note: This comprehensive list requires extensive traffic generator + capabilities. + + The match action(s) that were applied as part of the test should be + reported in the final test report. + + During this test, the DUT must perform the following operations on + the traffic flow: + + * Perform packet parsing on the DUT’s ingress port. + * Perform any relevant address look-ups on the DUT’s ingress + ports. + * Carry out one or more of the match actions specified above. + + The default loss percentages to be tested are: - X = 0% - X = 10^-7% + Other values can be tested if required by the user. The selected + frame sizes are those previously defined under + :ref:`default-test-parameters`. + + The test can also be used to determine the average latency of the + traffic when a match action is applied to packets in a flow. Under + the RFC2544 test methodology, the test duration will include a + number of trials; each trial should run for a minimum period of 60 + seconds. A binary search methodology must be applied for each + trial to obtain the final result. + + **Expected Result:** + + At the end of each trial, the presence or absence of loss + determines the modification of offered load for the next trial, + converging on a maximum rate, or RFC2544Throughput with X% loss. + The Throughput load is re-used in related RFC2544 tests and other + tests. + + **Metrics Collected:** + + The following are the metrics collected for this test: + + * The RFC 2544 Throughput in Frames Per Second (FPS) and Mbps + of the DUT for each frame size with X% packet loss. + * The average latency of the traffic flow when passing through + the DUT (if testing for latency, note that this average is + different from the test specified in Section 26.3 ofRFC2544). + * CPU and memory utilization may also be collected as part of + this test, to determine the vSwitch’s performance footprint + on the system. + + The metrics collected can be compared to that of the prerequisite + test to determine the cost of the match action(s) in the pipeline. + + **Deployment scenario**: + + - Physical → virtual switch → physical (and others are possible) + + [1] ovs-ofctl - administer OpenFlow switches + [http://openvswitch.org/support/dist-docs/ovs-ofctl.8.txt ] + + +.. 3.2.2.2 + +Packet Latency tests +-------------------- + +These tests will measure the store and forward latency as well as the packet +delay variation for various packet types through the virtual switch. The +following list is not exhaustive but should indicate the type of tests +that should be required. It is expected that more will be added. + +.. 3.2.2.2.1 + +Test ID: LTD.PacketLatency.InitialPacketProcessingLatency +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: Initial Packet Processing Latency + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + In some virtual switch architectures, the first packets of a flow will + take the system longer to process than subsequent packets in the flow. + This test determines the latency for these packets. The test will + measure the latency of the packets as they are processed by the + flow-setup-path of the DUT. There are two methods for this test, a + recommended method and a nalternative method that can be used if it is + possible to disable the fastpath of the virtual switch. + + Recommended method: This test will send 64,000 packets to the DUT, each + belonging to a different flow. Average packet latency will be determined + over the 64,000 packets. + + Alternative method: This test will send a single packet to the DUT after + a fixed interval of time. The time interval will be equivalent to the + amount of time it takes for a flow to time out in the virtual switch + plus 10%. Average packet latency will be determined over 1,000,000 + packets. + + This test is intended only for non-learning virtual switches; For learning + virtual switches use RFC2889. + + For this test, only unidirectional traffic is required. + + **Expected Result**: The average latency for the initial packet of all + flows should be greater than the latency of subsequent traffic. + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - Average latency of the initial packets of all flows that are + processed by the DUT. + + **Deployment scenario**: + + - Physical → Virtual Switch → Physical. + +.. 3.2.2.2.2 + +Test ID: LTD.PacketDelayVariation.RFC3393.Soak +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: Packet Delay Variation Soak Test + + **Prerequisite Tests**: LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss) + + **Priority**: + + **Description**: + + The aim of this test is to understand the distribution of packet delay + variation for different frame sizes over an extended test duration and + to determine if there are any outliers. To allow for an extended test + duration, the test should ideally run for 24 hours or, if this is not + possible, for at least 6 hour. For this test, each frame size must be + sent at the highest possible throughput with 0% packet loss, as + determined in the prerequisite test. + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The packet delay variation value for traffic passing through the DUT. + - The `RFC5481 `__ + PDV form of delay variation on the traffic flow, + using the 99th percentile, for each 60s interval during the test. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + +.. 3.2.2.3 + +Scalability tests +----------------- + +The general aim of these tests is to understand the impact of large flow +table size and flow lookups on throughput. The following list is not +exhaustive but should indicate the type of tests that should be required. +It is expected that more will be added. + +.. 3.2.2.3.1 + +.. _Scalability0PacketLoss: + +Test ID: LTD.Scalability.Flows.RFC2544.0PacketLoss +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 0% loss Flow Scalability throughput test + + **Prerequisite Test**: LTD.Throughput.RFC2544.PacketLossRatio, IF the + delta Throughput between the single-flow RFC2544 test and this test with + a variable number of flows is desired. + + **Priority**: + + **Description**: + + The aim of this test is to measure how throughput changes as the number + of flows in the DUT increases. The test will measure the throughput + through the fastpath, as such the flows need to be installed on the DUT + before passing traffic. + + For each frame size previously defined under :ref:`default-test-parameters` + and for each of the following number of flows: + + - 1,000 + - 2,000 + - 4,000 + - 8,000 + - 16,000 + - 32,000 + - 64,000 + - Max supported number of flows. + + This test will be conducted under two conditions following the + establishment of all flows as required by RFC 2544, regarding the flow + expiration time-out: + + 1) The time-out never expires during each trial. + + 2) The time-out expires for all flows periodically. This would require a + short time-out compared with flow re-appearance for a small number of + flows, and may not be possible for all flow conditions. + + The maximum 0% packet loss Throughput should be determined in a manner + identical to LTD.Throughput.RFC2544.PacketLossRatio. + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The maximum number of frames per second that can be forwarded at the + specified number of flows and the specified frame size, with zero + packet loss. + +.. 3.2.2.3.2 + +Test ID: LTD.MemoryBandwidth.RFC2544.0PacketLoss.Scalability +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 0% loss Memory Bandwidth Scalability test + + **Prerequisite Tests**: LTD.Throughput.RFC2544.PacketLossRatio, IF the + delta Throughput between an undisturbed RFC2544 test and this test with + the Throughput affected by cache and memory bandwidth contention is desired. + + **Priority**: + + **Description**: + + The aim of this test is to understand how the DUT's performance is + affected by cache sharing and memory bandwidth between processes. + + During the test all cores not used by the vSwitch should be running a + memory intensive application. This application should read and write + random data to random addresses in unused physical memory. The random + nature of the data and addresses is intended to consume cache, exercise + main memory access (as opposed to cache) and exercise all memory buses + equally. Furthermore: + + - the ratio of reads to writes should be recorded. A ratio of 1:1 + SHOULD be used. + - the reads and writes MUST be of cache-line size and be cache-line aligned. + - in NUMA architectures memory access SHOULD be local to the core's node. + Whether only local memory or a mix of local and remote memory is used + MUST be recorded. + - the memory bandwidth (reads plus writes) used per-core MUST be recorded; + the test MUST be run with a per-core memory bandwidth equal to half the + maximum system memory bandwidth divided by the number of cores. The test + MAY be run with other values for the per-core memory bandwidth. + - the test MAY also be run with the memory intensive application running + on all cores. + + Under these conditions the DUT's 0% packet loss throughput is determined + as per LTD.Throughput.RFC2544.PacketLossRatio. + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The DUT's 0% packet loss throughput in the presence of cache sharing and + memory bandwidth between processes. + +.. 3.2.2.3.3 + +Test ID: LTD.Scalability.VNF.RFC2544.PacketLossRatio +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: VNF Scalability RFC 2544 X% packet loss ratio Throughput and + Latency Test + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + This test determines the DUT's throughput rate with X% traffic loss for + a constant load (fixed length frames at a fixed interval time) when the + number of VNFs on the DUT increases. The default loss percentages + to be tested are: - X = 0% - X = 10^-7% . The minimum number of + VNFs to be tested are 3. + + Flow classification should be conducted with L2, L3 and L4 matching + to understand the matching and scaling capability of the vSwitch. The + matching fields which were used as part of the test should be reported + as part of the benchmark report. + + The vSwitch is responsible for forwarding frames between the VNFs + + The SUT (vSwitch and VNF daisy chain) operation should be validated + before running the test. This may be completed by running a burst or + continuous stream of traffic through the SUT to ensure proper operation + before a test. + + **Note**: The traffic rate used to validate SUT operation should be low + enough not to stress the SUT. + + **Note**: Other values can be tested if required by the user. + + **Note**: The same VNF should be used in the "daisy chain" formation. + Each addition of a VNF should be conducted in a new test setup (The DUT + is brought down, then the DUT is brought up again). An atlernative approach + would be to continue to add VNFs without bringing down the DUT. The + approach used needs to be documented as part of the test report. + + The selected frame sizes are those previously defined under + :ref:`default-test-parameters`. + The test can also be used to determine the average latency of the traffic. + + Under the `RFC2544 `__ + test methodology, the test duration will + include a number of trials; each trial should run for a minimum period + of 60 seconds. A binary search methodology must be applied for each + trial to obtain the final result for Throughput. + + **Expected Result**: At the end of each trial, the presence or absence + of loss determines the modification of offered load for the next trial, + converging on a maximum rate, or + `RFC2544 `__ Throughput with X% + loss. + The Throughput load is re-used in related + `RFC2544 `__ tests and other + tests. + + If the test VNFs are rather light-weight in terms of processing, the test + provides a view of multiple passes through the vswitch on logical + interfaces. In other words, the test produces an optimistic count of + daisy-chained VNFs, but the cumulative effect of traffic on the vSwitch is + "real" (assuming that the vSwitch has some dedicated resources, and the + effects on shared resources is understood). + + + **Metrics Collected**: + The following are the metrics collected for this test: + + - The maximum Throughput in Frames Per Second (FPS) and Mbps of + the DUT for each frame size with X% packet loss. + - The average latency of the traffic flow when passing through the DUT + and VNFs (if testing for latency, note that this average is different from the + test specified in Section 26.3 of + `RFC2544 `__). + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + +.. 3.2.2.3.4 + +Test ID: LTD.Scalability.VNF.RFC2544.PacketLossProfile +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: VNF Scalability RFC 2544 Throughput and Latency Profile + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + This test reveals how throughput and latency degrades as the number + of VNFs increases and offered rate varies in the region of the DUT's + maximum forwarding rate as determined by + LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss). + For example it can be used to determine if the degradation of throughput + and latency as the number of VNFs and offered rate increases is slow + and graceful, or sudden and severe. The minimum number of VNFs to + be tested is 3. + + The selected frame sizes are those previously defined under + :ref:`default-test-parameters`. + + The offered traffic rate is described as a percentage delta with respect + to the DUT's RFC 2544 Throughput as determined by + LTD.Throughput.RFC2544.PacketLoss Ratio (0% Packet Loss case). A delta + of 0% is equivalent to an offered traffic rate equal to the RFC 2544 + Throughput; A delta of +50% indicates an offered rate half-way + between the Throughput and line-rate, whereas a delta of + -50% indicates an offered rate of half the maximum rate. Therefore the + range of the delta figure is natuarlly bounded at -100% (zero offered + traffic) and +100% (traffic offered at line rate). + + The following deltas to the maximum forwarding rate should be applied: + + - -50%, -10%, 0%, +10% & +50% + + **Note**: Other values can be tested if required by the user. + + **Note**: The same VNF should be used in the "daisy chain" formation. + Each addition of a VNF should be conducted in a new test setup (The DUT + is brought down, then the DUT is brought up again). An atlernative approach + would be to continue to add VNFs without bringing down the DUT. The + approach used needs to be documented as part of the test report. + + Flow classification should be conducted with L2, L3 and L4 matching + to understand the matching and scaling capability of the vSwitch. The + matching fields which were used as part of the test should be reported + as part of the benchmark report. + + The SUT (vSwitch and VNF daisy chain) operation should be validated + before running the test. This may be completed by running a burst or + continuous stream of traffic through the SUT to ensure proper operation + before a test. + + **Note**: the traffic rate used to validate SUT operation should be low + enough not to stress the SUT + + **Expected Result**: For each packet size a profile should be produced + of how throughput and latency vary with offered rate. + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The forwarding rate in Frames Per Second (FPS) and Mbps of the DUT + for each delta to the maximum forwarding rate and for each frame + size. + - The average latency for each delta to the maximum forwarding rate and + for each frame size. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + - Any failures experienced (for example if the vSwitch crashes, stops + processing packets, restarts or becomes unresponsive to commands) + when the offered load is above Maximum Throughput MUST be recorded + and reported with the results. + +.. 3.2.2.4 + +Activation tests +---------------- + +The general aim of these tests is to understand the capacity of the +and speed with which the vswitch can accommodate new flows. + +.. 3.2.2.4.1 + +Test ID: LTD.Activation.RFC2889.AddressCachingCapacity +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2889 Address Caching Capacity Test + + **Prerequisite Test**: N/A + + **Priority**: + + **Description**: + + Please note this test is only applicable to virtual switches that are capable of + MAC learning. The aim of this test is to determine the address caching + capacity of the DUT for a constant load (fixed length frames at a fixed + interval time). The selected frame sizes are those previously defined + under :ref:`default-test-parameters`. + + In order to run this test the aging time, that is the maximum time the + DUT will keep a learned address in its flow table, and a set of initial + addresses, whose value should be >= 1 and <= the max number supported by + the implementation must be known. Please note that if the aging time is + configurable it must be longer than the time necessary to produce frames + from the external source at the specified rate. If the aging time is + fixed the frame rate must be brought down to a value that the external + source can produce in a time that is less than the aging time. + + Learning Frames should be sent from an external source to the DUT to + install a number of flows. The Learning Frames must have a fixed + destination address and must vary the source address of the frames. The + DUT should install flows in its flow table based on the varying source + addresses. Frames should then be transmitted from an external source at + a suitable frame rate to see if the DUT has properly learned all of the + addresses. If there is no frame loss and no flooding, the number of + addresses sent to the DUT should be increased and the test is repeated + until the max number of cached addresses supported by the DUT + determined. + + **Expected Result**: + + **Metrics collected**: + + The following are the metrics collected for this test: + + - Number of cached addresses supported by the DUT. + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + + **Deployment scenario**: + + - Physical → virtual switch → 2 x physical (one receiving, one listening). + +.. 3.2.2.4.2 + +Test ID: LTD.Activation.RFC2889.AddressLearningRate +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC2889 Address Learning Rate Test + + **Prerequisite Test**: LTD.Memory.RFC2889.AddressCachingCapacity + + **Priority**: + + **Description**: + + Please note this test is only applicable to virtual switches that are capable of + MAC learning. The aim of this test is to determine the rate of address + learning of the DUT for a constant load (fixed length frames at a fixed + interval time). The selected frame sizes are those previously defined + under :ref:`default-test-parameters`, traffic should be + sent with each IPv4/IPv6 address incremented by one. The rate at which + the DUT learns a new address should be measured. The maximum caching + capacity from LTD.Memory.RFC2889.AddressCachingCapacity should be taken + into consideration as the maximum number of addresses for which the + learning rate can be obtained. + + **Expected Result**: It may be worthwhile to report the behaviour when + operating beyond address capacity - some DUTs may be more friendly to + new addresses than others. + + **Metrics collected**: + + The following are the metrics collected for this test: + + - The address learning rate of the DUT. + + **Deployment scenario**: + + - Physical → virtual switch → 2 x physical (one receiving, one listening). + +.. 3.2.2.5 + +Coupling between control path and datapath Tests +------------------------------------------------ + +The following tests aim to determine how tightly coupled the datapath +and the control path are within a virtual switch. The following list +is not exhaustive but should indicate the type of tests that should be +required. It is expected that more will be added. + +.. 3.2.2.5.1 + +Test ID: LTD.CPDPCouplingFlowAddition +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: Control Path and Datapath Coupling + + **Prerequisite Test**: + + **Priority**: + + **Description**: + + The aim of this test is to understand how exercising the DUT's control + path affects datapath performance. + + Initially a certain number of flow table entries are installed in the + vSwitch. Then over the duration of an RFC2544 throughput test + flow-entries are added and removed at the rates specified below. No + traffic is 'hitting' these flow-entries, they are simply added and + removed. + + The test MUST be repeated with the following initial number of + flow-entries installed: - < 10 - 1000 - 100,000 - 10,000,000 (or the + maximum supported number of flow-entries) + + The test MUST be repeated with the following rates of flow-entry + addition and deletion per second: - 0 - 1 (i.e. 1 addition plus 1 + deletion) - 100 - 10,000 + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of + the DUT. + - The average latency of the traffic flow when passing through the DUT + (if testing for latency, note that this average is different from the + test specified in Section 26.3 of + `RFC2544 `__). + - CPU and memory utilization may also be collected as part of this + test, to determine the vSwitch's performance footprint on the system. + + **Deployment scenario**: + + - Physical → virtual switch → physical. + +.. 3.2.2.6 + +CPU and memory consumption +-------------------------- + +The following tests will profile a virtual switch's CPU and memory +utilization under various loads and circumstances. The following +list is not exhaustive but should indicate the type of tests that +should be required. It is expected that more will be added. + +.. 3.2.2.6.1 + +.. _CPU0PacketLoss: + +Test ID: LTD.Stress.RFC2544.0PacketLoss +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + **Title**: RFC 2544 0% Loss CPU OR Memory Stress Test + + **Prerequisite Test**: + + **Priority**: + + **Description**: + + The aim of this test is to understand the overall performance of the + system when a CPU or Memory intensive application is run on the same DUT as + the Virtual Switch. For each frame size, an + LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss) test should be + performed. Throughout the entire test a CPU or Memory intensive application + should be run on all cores on the system not in use by the Virtual Switch. + For NUMA system only cores on the same NUMA node are loaded. + + It is recommended that stress-ng be used for loading the non-Virtual + Switch cores but any stress tool MAY be used. + + **Expected Result**: + + **Metrics Collected**: + + The following are the metrics collected for this test: + + - Memory and CPU utilization of the cores running the Virtual Switch. + - The number of identity of the cores allocated to the Virtual Switch. + - The configuration of the stress tool (for example the command line + parameters used to start it.) + + **Note:** Stress in the test ID can be replaced with the name of the + component being stressed, when reporting the results: + LTD.CPU.RFC2544.0PacketLoss or LTD.Memory.RFC2544.0PacketLoss + +.. 3.2.2.7 + +Summary List of Tests +--------------------- + +1. Throughput tests + + - Test ID: LTD.Throughput.RFC2544.PacketLossRatio + - Test ID: LTD.Throughput.RFC2544.PacketLossRatioFrameModification + - Test ID: LTD.Throughput.RFC2544.Profile + - Test ID: LTD.Throughput.RFC2544.SystemRecoveryTime + - Test ID: LTD.Throughput.RFC2544.BackToBackFrames + - Test ID: LTD.Throughput.RFC2889.Soak + - Test ID: LTD.Throughput.RFC2889.SoakFrameModification + - Test ID: LTD.Throughput.RFC6201.ResetTime + - Test ID: LTD.Throughput.RFC2889.MaxForwardingRate + - Test ID: LTD.Throughput.RFC2889.ForwardPressure + - Test ID: LTD.Throughput.RFC2889.ErrorFramesFiltering + - Test ID: LTD.Throughput.RFC2889.BroadcastFrameForwarding + - Test ID: LTD.Throughput.RFC2544.WorstN-BestN + - Test ID: LTD.Throughput.Overlay.Network..RFC2544.PacketLossRatio + +2. Packet Latency tests + + - Test ID: LTD.PacketLatency.InitialPacketProcessingLatency + - Test ID: LTD.PacketDelayVariation.RFC3393.Soak + +3. Scalability tests + + - Test ID: LTD.Scalability.Flows.RFC2544.0PacketLoss + - Test ID: LTD.MemoryBandwidth.RFC2544.0PacketLoss.Scalability + - LTD.Scalability.VNF.RFC2544.PacketLossProfile + - LTD.Scalability.VNF.RFC2544.PacketLossRatio + +4. Activation tests + + - Test ID: LTD.Activation.RFC2889.AddressCachingCapacity + - Test ID: LTD.Activation.RFC2889.AddressLearningRate + +5. Coupling between control path and datapath Tests + + - Test ID: LTD.CPDPCouplingFlowAddition + +6. CPU and memory consumption + + - Test ID: LTD.Stress.RFC2544.0PacketLoss diff --git a/docs/testing/developer/requirements/vswitchperf_ltp.rst b/docs/testing/developer/requirements/vswitchperf_ltp.rst new file mode 100644 index 00000000..2b74d676 --- /dev/null +++ b/docs/testing/developer/requirements/vswitchperf_ltp.rst @@ -0,0 +1,1348 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intel Corporation, AT&T and others. + +.. 3.1 + +***************************** +VSPERF LEVEL TEST PLAN (LTP) +***************************** + +=============== +Introduction +=============== + +The objective of the OPNFV project titled +**Characterize vSwitch Performance for Telco NFV Use Cases**, is to +evaluate the performance of virtual switches to identify its suitability for a +Telco Network Function Virtualization (NFV) environment. The intention of this +Level Test Plan (LTP) document is to specify the scope, approach, resources, +and schedule of the virtual switch performance benchmarking activities in +OPNFV. The test cases will be identified in a separate document called the +Level Test Design (LTD) document. + +This document is currently in draft form. + +.. 3.1.1 + + +.. _doc-id: + +Document identifier +========================= + +The document id will be used to uniquely identify versions of the LTP. The +format for the document id will be: OPNFV\_vswitchperf\_LTP\_REL\_STATUS, where +by the status is one of: draft, reviewed, corrected or final. The document id +for this version of the LTP is: OPNFV\_vswitchperf\_LTP\_Colorado\_REVIEWED. + +.. 3.1.2 + +.. _scope: + +Scope +========== + +The main purpose of this project is to specify a suite of +performance tests in order to objectively measure the current packet +transfer characteristics of a virtual switch in the NFVI. The intent of +the project is to facilitate the performance testing of any virtual switch. +Thus, a generic suite of tests shall be developed, with no hard dependencies to +a single implementation. In addition, the test case suite shall be +architecture independent. + +The test cases developed in this project shall not form part of a +separate test framework, all of these tests may be inserted into the +Continuous Integration Test Framework and/or the Platform Functionality +Test Framework - if a vSwitch becomes a standard component of an OPNFV +release. + +.. 3.1.3 + +References +=============== + +* `RFC 1242 Benchmarking Terminology for Network Interconnection + Devices `__ +* `RFC 2544 Benchmarking Methodology for Network Interconnect + Devices `__ +* `RFC 2285 Benchmarking Terminology for LAN Switching + Devices `__ +* `RFC 2889 Benchmarking Methodology for LAN Switching + Devices `__ +* `RFC 3918 Methodology for IP Multicast + Benchmarking `__ +* `RFC 4737 Packet Reordering + Metrics `__ +* `RFC 5481 Packet Delay Variation Applicability + Statement `__ +* `RFC 6201 Device Reset + Characterization `__ + +.. 3.1.4 + +Level in the overall sequence +=============================== +The level of testing conducted by vswitchperf in the overall testing sequence (among +all the testing projects in OPNFV) is the performance benchmarking of a +specific component (the vswitch) in the OPNFV platfrom. It's expected that this +testing will follow on from the functional and integration testing conducted by +other testing projects in OPNFV, namely Functest and Yardstick. + +.. 3.1.5 + +Test classes and overall test conditions +========================================= +A benchmark is defined by the IETF as: A standardized test that serves as a +basis for performance evaluation and comparison. It's important to note that +benchmarks are not Functional tests. They do not provide PASS/FAIL criteria, +and most importantly ARE NOT performed on live networks, or performed with live +network traffic. + +In order to determine the packet transfer characteristics of a virtual switch, +the benchmarking tests will be broken down into the following categories: + +- **Throughput Tests** to measure the maximum forwarding rate (in + frames per second or fps) and bit rate (in Mbps) for a constant load + (as defined by `RFC1242 `__) + without traffic loss. +- **Packet and Frame Delay Tests** to measure average, min and max + packet and frame delay for constant loads. +- **Stream Performance Tests** (TCP, UDP) to measure bulk data transfer + performance, i.e. how fast systems can send and receive data through + the virtual switch. +- **Request/Response Performance** Tests (TCP, UDP) the measure the + transaction rate through the virtual switch. +- **Packet Delay Tests** to understand latency distribution for + different packet sizes and over an extended test run to uncover + outliers. +- **Scalability Tests** to understand how the virtual switch performs + as the number of flows, active ports, complexity of the forwarding + logic's configuration... it has to deal with increases. +- **Control Path and Datapath Coupling** Tests, to understand how + closely coupled the datapath and the control path are as well as the + effect of this coupling on the performance of the DUT. +- **CPU and Memory Consumption Tests** to understand the virtual + switch’s footprint on the system, this includes: + + * CPU core utilization. + * CPU cache utilization. + * Memory footprint. + * System bus (QPI, PCI, ..) utilization. + * Memory lanes utilization. + * CPU cycles consumed per packet. + * Time To Establish Flows Tests. + +- **Noisy Neighbour Tests**, to understand the effects of resource + sharing on the performance of a virtual switch. + +**Note:** some of the tests above can be conducted simultaneously where +the combined results would be insightful, for example Packet/Frame Delay +and Scalability. + + + +.. 3.2 + +.. _details-of-LTP: + +=================================== +Details of the Level Test Plan +=================================== + +This section describes the following items: +* Test items and their identifiers (TestItems_) +* Test Traceability Matrix (TestMatrix_) +* Features to be tested (FeaturesToBeTested_) +* Features not to be tested (FeaturesNotToBeTested_) +* Approach (Approach_) +* Item pass/fail criteria (PassFailCriteria_) +* Suspension criteria and resumption requirements (SuspensionResumptionReqs_) + +.. 3.2.1 + +.. _TestItems: + +Test items and their identifiers +================================== +The test item/application vsperf is trying to test are virtual switches and in +particular their performance in an nfv environment. vsperf will first try to +measure the maximum achievable performance by a virtual switch and then it will +focus in on usecases that are as close to real life deployment scenarios as +possible. + +.. 3.2.2 + +.. _TestMatrix: + +Test Traceability Matrix +========================== +vswitchperf leverages the "3x3" matrix (introduced in +https://tools.ietf.org/html/draft-ietf-bmwg-virtual-net-02) to achieve test +traceability. The matrix was expanded to 3x4 to accommodate scale metrics when +displaying the coverage of many metrics/benchmarks). Test case covreage in the +LTD is tracked using the following catagories: + + ++---------------+-------------+------------+---------------+-------------+ +| | | | | | +| | SPEED | ACCURACY | RELIABILITY | SCALE | +| | | | | | ++---------------+-------------+------------+---------------+-------------+ +| | | | | | +| Activation | X | X | X | X | +| | | | | | ++---------------+-------------+------------+---------------+-------------+ +| | | | | | +| Operation | X | X | X | X | +| | | | | | ++---------------+-------------+------------+---------------+-------------+ +| | | | | | +| De-activation | | | | | +| | | | | | ++---------------+-------------+------------+---------------+-------------+ + +X = denotes a test catagory that has 1 or more test cases defined. + +.. 3.2.3 + +.. _FeaturesToBeTested: + +Features to be tested +========================== + +Characterizing virtual switches (i.e. Device Under Test (DUT) in this document) +includes measuring the following performance metrics: + +- **Throughput** as defined by `RFC1242 + `__: The maximum rate at which + **none** of the offered frames are dropped by the DUT. The maximum frame + rate and bit rate that can be transmitted by the DUT without any error + should be recorded. Note there is an equivalent bit rate and a specific + layer at which the payloads contribute to the bits. Errors and + improperly formed frames or packets are dropped. +- **Packet delay** introduced by the DUT and its cumulative effect on + E2E networks. Frame delay can be measured equivalently. +- **Packet delay variation**: measured from the perspective of the + VNF/application. Packet delay variation is sometimes called "jitter". + However, we will avoid the term "jitter" as the term holds different + meaning to different groups of people. In this document we will + simply use the term packet delay variation. The preferred form for this + metric is the PDV form of delay variation defined in `RFC5481 + `__. The most relevant + measurement of PDV considers the delay variation of a single user flow, + as this will be relevant to the size of end-system buffers to compensate + for delay variation. The measurement system's ability to store the + delays of individual packets in the flow of interest is a key factor + that determines the specific measurement method. At the outset, it is + ideal to view the complete PDV distribution. Systems that can capture + and store packets and their delays have the freedom to calculate the + reference minimum delay and to determine various quantiles of the PDV + distribution accurately (in post-measurement processing routines). + Systems without storage must apply algorithms to calculate delay and + statistical measurements on the fly. For example, a system may store + temporary estimates of the mimimum delay and the set of (100) packets + with the longest delays during measurement (to calculate a high quantile, + and update these sets with new values periodically. + In some cases, a limited number of delay histogram bins will be + available, and the bin limits will need to be set using results from + repeated experiments. See section 8 of `RFC5481 + `__. +- **Packet loss** (within a configured waiting time at the receiver): All + packets sent to the DUT should be accounted for. +- **Burst behaviour**: measures the ability of the DUT to buffer packets. +- **Packet re-ordering**: measures the ability of the device under test to + maintain sending order throughout transfer to the destination. +- **Packet correctness**: packets or Frames must be well-formed, in that + they include all required fields, conform to length requirements, pass + integrity checks, etc. +- **Availability and capacity** of the DUT i.e. when the DUT is fully “up” + and connected, following measurements should be captured for + DUT without any network packet load: + + - Includes average power consumption of the CPUs (in various power states) and + system over specified period of time. Time period should not be less + than 60 seconds. + - Includes average per core CPU utilization over specified period of time. + Time period should not be less than 60 seconds. + - Includes the number of NIC interfaces supported. + - Includes headroom of VM workload processing cores (i.e. available + for applications). + +.. 3.2.4 + +.. _FeaturesNotToBeTested: + +Features not to be tested +========================== +vsperf doesn't intend to define or perform any functional tests. The aim is to +focus on performance. + +.. 3.2.5 + +.. _Approach: + +Approach +============== +The testing approach adoped by the vswitchperf project is black box testing, +meaning the test inputs can be generated and the outputs captured and +completely evaluated from the outside of the System Under Test. Some metrics +can be collected on the SUT, such as cpu or memory utilization if the +collection has no/minimal impact on benchmark. +This section will look at the deployment scenarios and the general methodology +used by vswitchperf. In addition, this section will also specify the details of +the Test Report that must be collected for each of the test cases. + +.. 3.2.5.1 + +Deployment Scenarios +-------------------------- +The following represents possible deployment test scenarios which can +help to determine the performance of both the virtual switch and the +datapaths to physical ports (to NICs) and to logical ports (to VNFs): + +.. 3.2.5.1.1 + +.. _Phy2Phy: + +Physical port → vSwitch → physical port +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +.. code-block:: console + + _ + +--------------------------------------------------+ | + | +--------------------+ | | + | | | | | + | | v | | Host + | +--------------+ +--------------+ | | + | | phy port | vSwitch | phy port | | | + +---+--------------+------------+--------------+---+ _| + ^ : + | | + : v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +.. 3.2.5.1.2 + +.. _PVP: + +Physical port → vSwitch → VNF → vSwitch → physical port +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +.. code-block:: console + + _ + +---------------------------------------------------+ | + | | | + | +-------------------------------------------+ | | + | | Application | | | + | +-------------------------------------------+ | | + | ^ : | | + | | | | | Guest + | : v | | + | +---------------+ +---------------+ | | + | | logical port 0| | logical port 1| | | + +---+---------------+-----------+---------------+---+ _| + ^ : + | | + : v _ + +---+---------------+----------+---------------+---+ | + | | logical port 0| | logical port 1| | | + | +---------------+ +---------------+ | | + | ^ : | | + | | | | | Host + | : v | | + | +--------------+ +--------------+ | | + | | phy port | vSwitch | phy port | | | + +---+--------------+------------+--------------+---+ _| + ^ : + | | + : v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +.. 3.2.5.1.3 + +.. _PVVP: + +Physical port → vSwitch → VNF → vSwitch → VNF → vSwitch → physical port +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +.. code-block:: console + + _ + +----------------------+ +----------------------+ | + | Guest 1 | | Guest 2 | | + | +---------------+ | | +---------------+ | | + | | Application | | | | Application | | | + | +---------------+ | | +---------------+ | | + | ^ | | | ^ | | | + | | v | | | v | | Guests + | +---------------+ | | +---------------+ | | + | | logical ports | | | | logical ports | | | + | | 0 1 | | | | 0 1 | | | + +---+---------------+--+ +---+---------------+--+ _| + ^ : ^ : + | | | | + : v : v _ + +---+---------------+---------+---------------+--+ | + | | 0 1 | | 3 4 | | | + | | logical ports | | logical ports | | | + | +---------------+ +---------------+ | | + | ^ | ^ | | | Host + | | L-----------------+ v | | + | +--------------+ +--------------+ | | + | | phy ports | vSwitch | phy ports | | | + +---+--------------+----------+--------------+---+ _| + ^ ^ : : + | | | | + : : v v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +.. 3.2.5.1.4 + +Physical port → VNF → vSwitch → VNF → physical port +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +.. code-block:: console + + _ + +----------------------+ +----------------------+ | + | Guest 1 | | Guest 2 | | + |+-------------------+ | | +-------------------+| | + || Application | | | | Application || | + |+-------------------+ | | +-------------------+| | + | ^ | | | ^ | | | Guests + | | v | | | v | | + |+-------------------+ | | +-------------------+| | + || logical ports | | | | logical ports || | + || 0 1 | | | | 0 1 || | + ++--------------------++ ++--------------------++ _| + ^ : ^ : + (PCI passthrough) | | (PCI passthrough) + | v : | _ + +--------++------------+-+------------++---------+ | + | | || 0 | | 1 || | | | + | | ||logical port| |logical port|| | | | + | | |+------------+ +------------+| | | | + | | | | ^ | | | | + | | | L-----------------+ | | | | + | | | | | | | Host + | | | vSwitch | | | | + | | +-----------------------------+ | | | + | | | | | + | | v | | + | +--------------+ +--------------+ | | + | | phy port/VF | | phy port/VF | | | + +-+--------------+--------------+--------------+-+ _| + ^ : + | | + : v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +.. 3.2.5.1.5 + +Physical port → vSwitch → VNF +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +.. code-block:: console + + _ + +---------------------------------------------------+ | + | | | + | +-------------------------------------------+ | | + | | Application | | | + | +-------------------------------------------+ | | + | ^ | | + | | | | Guest + | : | | + | +---------------+ | | + | | logical port 0| | | + +---+---------------+-------------------------------+ _| + ^ + | + : _ + +---+---------------+------------------------------+ | + | | logical port 0| | | + | +---------------+ | | + | ^ | | + | | | | Host + | : | | + | +--------------+ | | + | | phy port | vSwitch | | + +---+--------------+------------ -------------- ---+ _| + ^ + | + : + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +.. 3.2.5.1.6 + +VNF → vSwitch → physical port +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +.. code-block:: console + + _ + +---------------------------------------------------+ | + | | | + | +-------------------------------------------+ | | + | | Application | | | + | +-------------------------------------------+ | | + | : | | + | | | | Guest + | v | | + | +---------------+ | | + | | logical port | | | + +-------------------------------+---------------+---+ _| + : + | + v _ + +------------------------------+---------------+---+ | + | | logical port | | | + | +---------------+ | | + | : | | + | | | | Host + | v | | + | +--------------+ | | + | vSwitch | phy port | | | + +-------------------------------+--------------+---+ _| + : + | + v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +.. 3.2.5.1.7 + +VNF → vSwitch → VNF → vSwitch +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +.. code-block:: console + + _ + +-------------------------+ +-------------------------+ | + | Guest 1 | | Guest 2 | | + | +-----------------+ | | +-----------------+ | | + | | Application | | | | Application | | | + | +-----------------+ | | +-----------------+ | | + | : | | ^ | | + | | | | | | | Guest + | v | | : | | + | +---------------+ | | +---------------+ | | + | | logical port 0| | | | logical port 0| | | + +-----+---------------+---+ +---+---------------+-----+ _| + : ^ + | | + v : _ + +----+---------------+------------+---------------+-----+ | + | | port 0 | | port 1 | | | + | +---------------+ +---------------+ | | + | : ^ | | + | | | | | Host + | +--------------------+ | | + | | | + | vswitch | | + +-------------------------------------------------------+ _| + +.. 3.2.5.1.8 + +HOST 1(Physical port → virtual switch → VNF → virtual switch → Physical port) +→ HOST 2(Physical port → virtual switch → VNF → virtual switch → Physical port) + +HOST 1 (PVP) → HOST 2 (PVP) +~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +.. code-block:: console + + _ + +----------------------+ +----------------------+ | + | Guest 1 | | Guest 2 | | + | +---------------+ | | +---------------+ | | + | | Application | | | | Application | | | + | +---------------+ | | +---------------+ | | + | ^ | | | ^ | | | + | | v | | | v | | Guests + | +---------------+ | | +---------------+ | | + | | logical ports | | | | logical ports | | | + | | 0 1 | | | | 0 1 | | | + +---+---------------+--+ +---+---------------+--+ _| + ^ : ^ : + | | | | + : v : v _ + +---+---------------+--+ +---+---------------+--+ | + | | 0 1 | | | | 3 4 | | | + | | logical ports | | | | logical ports | | | + | +---------------+ | | +---------------+ | | + | ^ | | | ^ | | | Hosts + | | v | | | v | | + | +--------------+ | | +--------------+ | | + | | phy ports | | | | phy ports | | | + +---+--------------+---+ +---+--------------+---+ _| + ^ : : : + | +-----------------+ | + : v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + + + +**Note:** For tests where the traffic generator and/or measurement +receiver are implemented on VM and connected to the virtual switch +through vNIC, the issues of shared resources and interactions between +the measurement devices and the device under test must be considered. + +**Note:** Some RFC 2889 tests require a full-mesh sending and receiving +pattern involving more than two ports. This possibility is illustrated in the +Physical port → vSwitch → VNF → vSwitch → VNF → vSwitch → physical port +diagram above (with 2 sending and 2 receiving ports, though all ports +could be used bi-directionally). + +**Note:** When Deployment Scenarios are used in RFC 2889 address learning +or cache capacity testing, an additional port from the vSwitch must be +connected to the test device. This port is used to listen for flooded +frames. + +.. 3.2.5.2 + +General Methodology: +-------------------------- +To establish the baseline performance of the virtual switch, tests would +initially be run with a simple workload in the VNF (the recommended +simple workload VNF would be `DPDK `__'s testpmd +application forwarding packets in a VM or vloop\_vnf a simple kernel +module that forwards traffic between two network interfaces inside the +virtualized environment while bypassing the networking stack). +Subsequently, the tests would also be executed with a real Telco +workload running in the VNF, which would exercise the virtual switch in +the context of higher level Telco NFV use cases, and prove that its +underlying characteristics and behaviour can be measured and validated. +Suitable real Telco workload VNFs are yet to be identified. + +.. 3.2.5.2.1 + +.. _default-test-parameters: + +Default Test Parameters +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The following list identifies the default parameters for suite of +tests: + +- Reference application: Simple forwarding or Open Source VNF. +- Frame size (bytes): 64, 128, 256, 512, 1024, 1280, 1518, 2K, 4k OR + Packet size based on use-case (e.g. RTP 64B, 256B) OR Mix of packet sizes as + maintained by the Functest project . +- Reordering check: Tests should confirm that packets within a flow are + not reordered. +- Duplex: Unidirectional / Bidirectional. Default: Full duplex with + traffic transmitting in both directions, as network traffic generally + does not flow in a single direction. By default the data rate of + transmitted traffic should be the same in both directions, please + note that asymmetric traffic (e.g. downlink-heavy) tests will be + mentioned explicitly for the relevant test cases. +- Number of Flows: Default for non scalability tests is a single flow. + For scalability tests the goal is to test with maximum supported + flows but where possible will test up to 10 Million flows. Start with + a single flow and scale up. By default flows should be added + sequentially, tests that add flows simultaneously will explicitly + call out their flow addition behaviour. Packets are generated across + the flows uniformly with no burstiness. For multi-core tests should + consider the number of packet flows based on vSwitch/VNF multi-thread + implementation and behavior. + +- Traffic Types: UDP, SCTP, RTP, GTP and UDP traffic. +- Deployment scenarios are: +- Physical → virtual switch → physical. +- Physical → virtual switch → VNF → virtual switch → physical. +- Physical → virtual switch → VNF → virtual switch → VNF → virtual + switch → physical. +- Physical → VNF → virtual switch → VNF → physical. +- Physical → virtual switch → VNF. +- VNF → virtual switch → Physical. +- VNF → virtual switch → VNF. + +Tests MUST have these parameters unless otherwise stated. **Test cases +with non default parameters will be stated explicitly**. + +**Note**: For throughput tests unless stated otherwise, test +configurations should ensure that traffic traverses the installed flows +through the virtual switch, i.e. flows are installed and have an appropriate +time out that doesn't expire before packet transmission starts. + +.. 3.2.5.2.2 + +Flow Classification +~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +Virtual switches classify packets into flows by processing and matching +particular header fields in the packet/frame and/or the input port where +the packets/frames arrived. The vSwitch then carries out an action on +the group of packets that match the classification parameters. Thus a +flow is considered to be a sequence of packets that have a shared set of +header field values or have arrived on the same port and have the same +action applied to them. Performance results can vary based on the +parameters the vSwitch uses to match for a flow. The recommended flow +classification parameters for L3 vSwitch performance tests are: the +input port, the source IP address, the destination IP address and the +Ethernet protocol type field. It is essential to increase the flow +time-out time on a vSwitch before conducting any performance tests that +do not measure the flow set-up time. Normally the first packet of a +particular flow will install the flow in the vSwitch which adds an +additional latency, subsequent packets of the same flow are not subject +to this latency if the flow is already installed on the vSwitch. + +.. 3.2.5.2.3 + +Test Priority +~~~~~~~~~~~~~~~~~~~~~ + +Tests will be assigned a priority in order to determine which tests +should be implemented immediately and which tests implementations +can be deferred. + +Priority can be of following types: - Urgent: Must be implemented +immediately. - High: Must be implemented in the next release. - Medium: +May be implemented after the release. - Low: May or may not be +implemented at all. + +.. 3.2.5.2.4 + +SUT Setup +~~~~~~~~~~~~~~~~~~ + +The SUT should be configured to its "default" state. The +SUT's configuration or set-up must not change between tests in any way +other than what is required to do the test. All supported protocols must +be configured and enabled for each test set up. + +.. 3.2.5.2.5 + +Port Configuration +~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The DUT should be configured with n ports where +n is a multiple of 2. Half of the ports on the DUT should be used as +ingress ports and the other half of the ports on the DUT should be used +as egress ports. Where a DUT has more than 2 ports, the ingress data +streams should be set-up so that they transmit packets to the egress +ports in sequence so that there is an even distribution of traffic +across ports. For example, if a DUT has 4 ports 0(ingress), 1(ingress), +2(egress) and 3(egress), the traffic stream directed at port 0 should +output a packet to port 2 followed by a packet to port 3. The traffic +stream directed at port 1 should also output a packet to port 2 followed +by a packet to port 3. + +.. 3.2.5.2.6 + +Frame Formats +~~~~~~~~~~~~~~~~~~~~~ + +**Frame formats Layer 2 (data link layer) protocols** + +- Ethernet II + +.. code-block:: console + + +---------------------------+-----------+ + | Ethernet Header | Payload | Check Sum | + +-----------------+---------+-----------+ + |_________________|_________|___________| + 14 Bytes 46 - 1500 4 Bytes + Bytes + + +**Layer 3 (network layer) protocols** + +- IPv4 + +.. code-block:: console + + +-----------------+-----------+---------+-----------+ + | Ethernet Header | IP Header | Payload | Checksum | + +-----------------+-----------+---------+-----------+ + |_________________|___________|_________|___________| + 14 Bytes 20 bytes 26 - 1480 4 Bytes + Bytes + +- IPv6 + +.. code-block:: console + + +-----------------+-----------+---------+-----------+ + | Ethernet Header | IP Header | Payload | Checksum | + +-----------------+-----------+---------+-----------+ + |_________________|___________|_________|___________| + 14 Bytes 40 bytes 26 - 1460 4 Bytes + Bytes + +**Layer 4 (transport layer) protocols** + + - TCP + - UDP + - SCTP + +.. code-block:: console + + +-----------------+-----------+-----------------+---------+-----------+ + | Ethernet Header | IP Header | Layer 4 Header | Payload | Checksum | + +-----------------+-----------+-----------------+---------+-----------+ + |_________________|___________|_________________|_________|___________| + 14 Bytes 40 bytes 20 Bytes 6 - 1460 4 Bytes + Bytes + + +**Layer 5 (application layer) protocols** + + - RTP + - GTP + +.. code-block:: console + + +-----------------+-----------+-----------------+---------+-----------+ + | Ethernet Header | IP Header | Layer 4 Header | Payload | Checksum | + +-----------------+-----------+-----------------+---------+-----------+ + |_________________|___________|_________________|_________|___________| + 14 Bytes 20 bytes 20 Bytes >= 6 Bytes 4 Bytes + +.. 3.2.5.2.7 + +Packet Throughput +~~~~~~~~~~~~~~~~~~~~~~~~~ +There is a difference between an Ethernet frame, +an IP packet, and a UDP datagram. In the seven-layer OSI model of +computer networking, packet refers to a data unit at layer 3 (network +layer). The correct term for a data unit at layer 2 (data link layer) is +a frame, and at layer 4 (transport layer) is a segment or datagram. + +Important concepts related to 10GbE performance are frame rate and +throughput. The MAC bit rate of 10GbE, defined in the IEEE standard 802 +.3ae, is 10 billion bits per second. Frame rate is based on the bit rate +and frame format definitions. Throughput, defined in IETF RFC 1242, is +the highest rate at which the system under test can forward the offered +load, without loss. + +The frame rate for 10GbE is determined by a formula that divides the 10 +billion bits per second by the preamble + frame length + inter-frame +gap. + +The maximum frame rate is calculated using the minimum values of the +following parameters, as described in the IEEE 802 .3ae standard: + +- Preamble: 8 bytes \* 8 = 64 bits +- Frame Length: 64 bytes (minimum) \* 8 = 512 bits +- Inter-frame Gap: 12 bytes (minimum) \* 8 = 96 bits + +Therefore, Maximum Frame Rate (64B Frames) += MAC Transmit Bit Rate / (Preamble + Frame Length + Inter-frame Gap) += 10,000,000,000 / (64 + 512 + 96) += 10,000,000,000 / 672 += 14,880,952.38 frame per second (fps) + +.. 3.2.5.3 + +RFCs for testing virtual switch performance +-------------------------------------------------- + +The starting point for defining the suite of tests for benchmarking the +performance of a virtual switch is to take existing RFCs and standards +that were designed to test their physical counterparts and adapting them +for testing virtual switches. The rationale behind this is to establish +a fair comparison between the performance of virtual and physical +switches. This section outlines the RFCs that are used by this +specification. + +.. 3.2.5.3.1 + +RFC 1242 Benchmarking Terminology for Network Interconnection +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +Devices RFC 1242 defines the terminology that is used in describing +performance benchmarking tests and their results. Definitions and +discussions covered include: Back-to-back, bridge, bridge/router, +constant load, data link frame size, frame loss rate, inter frame gap, +latency, and many more. + +.. 3.2.5.3.2 + +RFC 2544 Benchmarking Methodology for Network Interconnect Devices +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 2544 outlines a benchmarking methodology for network Interconnect +Devices. The methodology results in performance metrics such as latency, +frame loss percentage, and maximum data throughput. + +In this document network “throughput” (measured in millions of frames +per second) is based on RFC 2544, unless otherwise noted. Frame size +refers to Ethernet frames ranging from smallest frames of 64 bytes to +largest frames of 9K bytes. + +Types of tests are: + +1. Throughput test defines the maximum number of frames per second + that can be transmitted without any error, or 0% loss ratio. + In some Throughput tests (and those tests with long duration), + evaluation of an additional frame loss ratio is suggested. The + current ratio (10^-7 %) is based on understanding the typical + user-to-user packet loss ratio needed for good application + performance and recognizing that a single transfer through a + vswitch must contribute a tiny fraction of user-to-user loss. + Further, the ratio 10^-7 % also recognizes practical limitations + when measuring loss ratio. + +2. Latency test measures the time required for a frame to travel from + the originating device through the network to the destination device. + Please note that RFC2544 Latency measurement will be superseded with + a measurement of average latency over all successfully transferred + packets or frames. + +3. Frame loss test measures the network’s + response in overload conditions - a critical indicator of the + network’s ability to support real-time applications in which a + large amount of frame loss will rapidly degrade service quality. + +4. Burst test assesses the buffering capability of a virtual switch. It + measures the maximum number of frames received at full line rate + before a frame is lost. In carrier Ethernet networks, this + measurement validates the excess information rate (EIR) as defined in + many SLAs. + +5. System recovery to characterize speed of recovery from an overload + condition. + +6. Reset to characterize speed of recovery from device or software + reset. This type of test has been updated by `RFC6201 + `__ as such, + the methodology defined by this specification will be that of RFC 6201. + +Although not included in the defined RFC 2544 standard, another crucial +measurement in Ethernet networking is packet delay variation. The +definition set out by this specification comes from +`RFC5481 `__. + +.. 3.2.5.3.3 + +RFC 2285 Benchmarking Terminology for LAN Switching Devices +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 2285 defines the terminology that is used to describe the +terminology for benchmarking a LAN switching device. It extends RFC +1242 and defines: DUTs, SUTs, Traffic orientation and distribution, +bursts, loads, forwarding rates, etc. + +.. 3.2.5.3.4 + +RFC 2889 Benchmarking Methodology for LAN Switching +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 2889 outlines a benchmarking methodology for LAN switching, it +extends RFC 2544. The outlined methodology gathers performance +metrics for forwarding, congestion control, latency, address handling +and finally filtering. + +.. 3.2.5.3.5 + +RFC 3918 Methodology for IP Multicast Benchmarking +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 3918 outlines a methodology for IP Multicast benchmarking. + +.. 3.2.5.3.6 + +RFC 4737 Packet Reordering Metrics +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 4737 describes metrics for identifying and counting re-ordered +packets within a stream, and metrics to measure the extent each +packet has been re-ordered. + +.. 3.2.5.3.7 + +RFC 5481 Packet Delay Variation Applicability Statement +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 5481 defined two common, but different forms of delay variation +metrics, and compares the metrics over a range of networking +circumstances and tasks. The most suitable form for vSwitch +benchmarking is the "PDV" form. + +.. 3.2.5.3.8 + +RFC 6201 Device Reset Characterization +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +RFC 6201 extends the methodology for characterizing the speed of +recovery of the DUT from device or software reset described in RFC +2544. + +.. 3.2.6: + +.. _PassFailCriteria: + +Item pass/fail criteria +========================= + +vswitchperf does not specify Pass/Fail criteria for the tests in terms of a +threshold, as benchmarks do not (and should not do this). The results/metrics +for a test are simply reported. If it had to be defined, a test is considered +to have passed if it succesfully completed and a relavent metric was +recorded/reported for the SUT. + +.. 3.2.7: + +.. _SuspensionResumptionReqs: + +Suspension criteria and resumption requirements +================================================ +In the case of a throughput test, a test should be suspended if a virtual +switch is failing to forward any traffic. A test should be restarted from a +clean state if the intention is to carry out the test again. + +.. 3.2.8: + +.. _TestDelierables: + +Test deliverables +================== +Each test should produce a test report that details SUT information as well as +the test results. There are a number of parameters related to the system, DUT +and tests that can affect the repeatability of a test results and should be +recorded. In order to minimise the variation in the results of a test, +it is recommended that the test report includes the following information: + +- Hardware details including: + + - Platform details. + - Processor details. + - Memory information (see below) + - Number of enabled cores. + - Number of cores used for the test. + - Number of physical NICs, as well as their details (manufacturer, + versions, type and the PCI slot they are plugged into). + - NIC interrupt configuration. + - BIOS version, release date and any configurations that were + modified. + +- Software details including: + + - OS version (for host and VNF) + - Kernel version (for host and VNF) + - GRUB boot parameters (for host and VNF). + - Hypervisor details (Type and version). + - Selected vSwitch, version number or commit id used. + - vSwitch launch command line if it has been parameterised. + - Memory allocation to the vSwitch – which NUMA node it is using, + and how many memory channels. + - Where the vswitch is built from source: compiler details including + versions and the flags that were used to compile the vSwitch. + - DPDK or any other SW dependency version number or commit id used. + - Memory allocation to a VM - if it's from Hugpages/elsewhere. + - VM storage type: snapshot/independent persistent/independent + non-persistent. + - Number of VMs. + - Number of Virtual NICs (vNICs), versions, type and driver. + - Number of virtual CPUs and their core affinity on the host. + - Number vNIC interrupt configuration. + - Thread affinitization for the applications (including the vSwitch + itself) on the host. + - Details of Resource isolation, such as CPUs designated for + Host/Kernel (isolcpu) and CPUs designated for specific processes + (taskset). + +- Memory Details + + - Total memory + - Type of memory + - Used memory + - Active memory + - Inactive memory + - Free memory + - Buffer memory + - Swap cache + - Total swap + - Used swap + - Free swap + +- Test duration. +- Number of flows. +- Traffic Information: + + - Traffic type - UDP, TCP, IMIX / Other. + - Packet Sizes. + +- Deployment Scenario. + +**Note**: Tests that require additional parameters to be recorded will +explicitly specify this. + + +.. 3.3: + +.. _TestManagement: + +Test management +================= +This section will detail the test activities that will be conducted by vsperf +as well as the infrastructure that will be used to complete the tests in OPNFV. + +.. 3.3.1: + +Planned activities and tasks; test progression +================================================= +A key consideration when conducting any sort of benchmark is trying to +ensure the consistency and repeatability of test results between runs. +When benchmarking the performance of a virtual switch there are many +factors that can affect the consistency of results. This section +describes these factors and the measures that can be taken to limit +their effects. In addition, this section will outline some system tests +to validate the platform and the VNF before conducting any vSwitch +benchmarking tests. + +**System Isolation:** + +When conducting a benchmarking test on any SUT, it is essential to limit +(and if reasonable, eliminate) any noise that may interfere with the +accuracy of the metrics collected by the test. This noise may be +introduced by other hardware or software (OS, other applications), and +can result in significantly varying performance metrics being collected +between consecutive runs of the same test. In the case of characterizing +the performance of a virtual switch, there are a number of configuration +parameters that can help increase the repeatability and stability of +test results, including: + +- OS/GRUB configuration: + + - maxcpus = n where n >= 0; limits the kernel to using 'n' + processors. Only use exactly what you need. + - isolcpus: Isolate CPUs from the general scheduler. Isolate all + CPUs bar one which will be used by the OS. + - use taskset to affinitize the forwarding application and the VNFs + onto isolated cores. VNFs and the vSwitch should be allocated + their own cores, i.e. must not share the same cores. vCPUs for the + VNF should be affinitized to individual cores also. + - Limit the amount of background applications that are running and + set OS to boot to runlevel 3. Make sure to kill any unnecessary + system processes/daemons. + - Only enable hardware that you need to use for your test – to + ensure there are no other interrupts on the system. + - Configure NIC interrupts to only use the cores that are not + allocated to any other process (VNF/vSwitch). + +- NUMA configuration: Any unused sockets in a multi-socket system + should be disabled. +- CPU pinning: The vSwitch and the VNF should each be affinitized to + separate logical cores using a combination of maxcpus, isolcpus and + taskset. +- BIOS configuration: BIOS should be configured for performance where + an explicit option exists, sleep states should be disabled, any + virtualization optimization technologies should be enabled, and + hyperthreading should also be enabled, turbo boost and overclocking + should be disabled. + +**System Validation:** + +System validation is broken down into two sub-categories: Platform +validation and VNF validation. The validation test itself involves +verifying the forwarding capability and stability for the sub-system +under test. The rationale behind system validation is two fold. Firstly +to give a tester confidence in the stability of the platform or VNF that +is being tested; and secondly to provide base performance comparison +points to understand the overhead introduced by the virtual switch. + +* Benchmark platform forwarding capability: This is an OPTIONAL test + used to verify the platform and measure the base performance (maximum + forwarding rate in fps and latency) that can be achieved by the + platform without a vSwitch or a VNF. The following diagram outlines + the set-up for benchmarking Platform forwarding capability: + + .. code-block:: console + + __ + +--------------------------------------------------+ | + | +------------------------------------------+ | | + | | | | | + | | l2fw or DPDK L2FWD app | | Host + | | | | | + | +------------------------------------------+ | | + | | NIC | | | + +---+------------------------------------------+---+ __| + ^ : + | | + : v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + +* Benchmark VNF forwarding capability: This test is used to verify + the VNF and measure the base performance (maximum forwarding rate in + fps and latency) that can be achieved by the VNF without a vSwitch. + The performance metrics collected by this test will serve as a key + comparison point for NIC passthrough technologies and vSwitches. VNF + in this context refers to the hypervisor and the VM. The following + diagram outlines the set-up for benchmarking VNF forwarding + capability: + + .. code-block:: console + + __ + +--------------------------------------------------+ | + | +------------------------------------------+ | | + | | | | | + | | VNF | | | + | | | | | + | +------------------------------------------+ | | + | | Passthrough/SR-IOV | | Host + | +------------------------------------------+ | | + | | NIC | | | + +---+------------------------------------------+---+ __| + ^ : + | | + : v + +--------------------------------------------------+ + | | + | traffic generator | + | | + +--------------------------------------------------+ + + +**Methodology to benchmark Platform/VNF forwarding capability** + + +The recommended methodology for the platform/VNF validation and +benchmark is: - Run `RFC2889 `__ +Maximum Forwarding Rate test, this test will produce maximum +forwarding rate and latency results that will serve as the +expected values. These expected values can be used in +subsequent steps or compared with in subsequent validation tests. - +Transmit bidirectional traffic at line rate/max forwarding rate +(whichever is higher) for at least 72 hours, measure throughput (fps) +and latency. - Note: Traffic should be bidirectional. - Establish a +baseline forwarding rate for what the platform can achieve. - Additional +validation: After the test has completed for 72 hours run bidirectional +traffic at the maximum forwarding rate once more to see if the system is +still functional and measure throughput (fps) and latency. Compare the +measure the new obtained values with the expected values. + +**NOTE 1**: How the Platform is configured for its forwarding capability +test (BIOS settings, GRUB configuration, runlevel...) is how the +platform should be configured for every test after this + +**NOTE 2**: How the VNF is configured for its forwarding capability test +(# of vCPUs, vNICs, Memory, affinitization…) is how it should be +configured for every test that uses a VNF after this. + +**Methodology to benchmark the VNF to vSwitch to VNF deployment scenario** + +vsperf has identified the following concerns when benchmarking the VNF to +vSwitch to VNF deployment scenario: + +* The accuracy of the timing synchronization between VNFs/VMs. +* The clock accuracy of a VNF/VM if they were to be used as traffic generators. +* VNF traffic generator/receiver may be using resources of the system under + test, causing at least three forms of workload to increase as the traffic + load increases (generation, switching, receiving). + +The recommendation from vsperf is that tests for this sceanario must +include an external HW traffic generator to act as the tester/traffic transmitter +and receiver. The perscribed methodology to benchmark this deployment scanrio with +an external tester involves the following three steps: + +#. Determine the forwarding capability and latency through the virtual interface +connected to the VNF/VM. + +.. Figure:: vm2vm_virtual_interface_benchmark.png + + Virtual interfaces performance benchmark + +#. Determine the forwarding capability and latency through the VNF/hypervisor. + +.. Figure:: vm2vm_hypervisor_benchmark.png + + Hypervisor performance benchmark + +#. Determine the forwarding capability and latency for the VNF to vSwitch to VNF + taking the information from the previous two steps into account. + +.. Figure:: vm2vm_benchmark.png + + VNF to vSwitch to VNF performance benchmark + +vsperf also identified an alternative configuration for the final step: + +.. Figure:: vm2vm_alternative_benchmark.png + + VNF to vSwitch to VNF alternative performance benchmark + +.. 3.3.2: + +Environment/infrastructure +============================ +Intel is providing a hosted test-bed with nine bare-metal environments +allocated to different OPNFV projects. Currently a number of servers in +`Intel POD 3 `__ are +allocated to vsperf: + + * pod3-wcp-node3 and pod3-wcp-node4 which are used for CI jobs. + * pod3-node6 which is used as a vsperf sandbox environment. + +vsperf CI +--------- +vsperf CI jobs are broken down into: + + * Daily job: + + * Runs everyday takes about 10 hours to complete. + * TESTCASES_DAILY='phy2phy_tput back2back phy2phy_tput_mod_vlan + phy2phy_scalability pvp_tput pvp_back2back pvvp_tput pvvp_back2back'. + * TESTPARAM_DAILY='--test-params TRAFFICGEN_PKT_SIZES=(64,128,512,1024,1518)'. + + * Merge job: + + * Runs whenever patches are merged to master. + * Runs a basic Sanity test. + + * Verify job: + + * Runs every time a patch is pushed to gerrit. + * Builds documentation. + +Scripts: +-------- +There are 2 scripts that are part of VSPERFs CI: + + * build-vsperf.sh: Lives in the VSPERF repository in the ci/ directory and is + used to run vsperf with the appropriate cli parameters. + * vswitchperf.yml: YAML description of our jenkins job. lives in the RELENG + repository. + +More info on vsperf CI can be found here: +https://wiki.opnfv.org/display/vsperf/VSPERF+CI + +.. 3.3.3: + +Responsibilities and authority +=============================== +The group responsible for managing, designing, preparing and executing the +tests listed in the LTD are the vsperf committers and contributors. The vsperf +committers and contributors should work with the relavent OPNFV projects to +ensure that the infrastructure is in place for testing vswitches, and that the +results are published to common end point (a results database). + diff --git a/docs/testing/developer/results/results.rst b/docs/testing/developer/results/results.rst new file mode 100644 index 00000000..df9c52cb --- /dev/null +++ b/docs/testing/developer/results/results.rst @@ -0,0 +1,38 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intel Corporation, AT&T and others. + +OPNFV Test Results +========================= +VSPERF CI jobs are run daily and sample results can be found at +https://wiki.opnfv.org/display/vsperf/Vsperf+Results + +The following example maps the results in the test dashboard to the appropriate +test case in the VSPERF Framework and specifies the metric the vertical/Y axis +is plotting. **Please note**, the presence of dpdk within a test name signifies +that the vswitch under test was OVS with DPDK, while its absence indicates that +the vswitch under test was stock OVS. + +===================== ===================== ================== =============== + Dashboard Test Framework Test Metric Guest Interface +===================== ===================== ================== =============== +tput_ovsdpdk phy2phy_tput Throughput (FPS) N/A +tput_ovs phy2phy_tput Throughput (FPS) N/A +b2b_ovsdpdk back2back Back-to-back value N/A +b2b_ovs back2back Back-to-back value N/A +tput_mod_vlan_ovs phy2phy_tput_mod_vlan Throughput (FPS) N/A +tput_mod_vlan_ovsdpdk phy2phy_tput_mod_vlan Throughput (FPS) N/A +scalability_ovs phy2phy_scalability Throughput (FPS) N/A +scalability_ovsdpdk phy2phy_scalability Throughput (FPS) N/A +pvp_tput_ovsdpdkuser pvp_tput Throughput (FPS) vhost-user +pvp_tput_ovsvirtio pvp_tput Throughput (FPS) virtio-net +pvp_b2b_ovsdpdkuser pvp_back2back Back-to-back value vhost-user +pvp_b2b_ovsvirtio pvp_back2back Back-to-back value virtio-net +pvvp_tput_ovsdpdkuser pvvp_tput Throughput (FPS) vhost-user +pvvp_tput_ovsvirtio pvvp_tput Throughput (FPS) virtio-net +pvvp_b2b_ovsdpdkuser pvvp_back2back Throughput (FPS) vhost-user +pvvp_b2b_ovsvirtio pvvp_back2back Throughput (FPS) virtio-net +===================== ===================== ================== =============== + +The loopback application in the VNF was used for PVP and PVVP scenarios was DPDK +testpmd. diff --git a/docs/testing/developer/results/scenario.rst b/docs/testing/developer/results/scenario.rst new file mode 100644 index 00000000..a57d6a03 --- /dev/null +++ b/docs/testing/developer/results/scenario.rst @@ -0,0 +1,47 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intel Corporation, AT&T and others. + +VSPERF Test Scenarios +===================== + +Predefined Tests run with CI: + +===================== =========================================================== + Test Definition +===================== =========================================================== +phy2phy_tput :ref:`PacketLossRatio ` for :ref:`Phy2Phy ` +back2back :ref:`BackToBackFrames ` for :ref:`Phy2Phy ` +phy2phy_tput_mod_vlan :ref:`PacketLossRatioFrameModification ` for :ref:`Phy2Phy ` +phy2phy_cont :ref:`Phy2Phy ` blast vswitch at x% TX rate and measure throughput +pvp_cont :ref:`PVP ` blast vswitch at x% TX rate and measure throughput +pvvp_cont :ref:`PVVP ` blast vswitch at x% TX rate and measure throughput +phy2phy_scalability :ref:`Scalability0PacketLoss ` for :ref:`Phy2Phy ` +pvp_tput :ref:`PacketLossRatio ` for :ref:`PVP ` +pvp_back2back :ref:`BackToBackFrames ` for :ref:`PVP ` +pvvp_tput :ref:`PacketLossRatio ` for :ref:`PVVP ` +pvvp_back2back :ref:`BackToBackFrames ` for :ref:`PVVP ` +phy2phy_cpu_load :ref:`CPU0PacketLoss ` for :ref:`Phy2Phy ` +phy2phy_mem_load Same as :ref:`CPU0PacketLoss ` but using a memory intensive app +===================== =========================================================== + +Deployment topologies: + +* :ref:`Phy2Phy `: Physical port -> vSwitch -> Physical port. +* :ref:`PVP `: Physical port -> vSwitch -> VNF -> vSwitch -> Physical port. +* :ref:`PVVP `: Physical port -> vSwitch -> VNF -> vSwitch -> VNF -> vSwitch -> + Physical port. + +Loopback applications in the Guest: + +* `DPDK testpmd `_. +* Linux Bridge. +* :ref:`l2fwd-module` + +Supported traffic generators: + +* Spirent Testcenter +* Ixia: IxOS and IxNet. +* Xena +* MoonGen +* Dummy -- cgit 1.2.3-korg