diff options
Diffstat (limited to 'docs/testing/user/userguide/12-nsb-overview.rst')
| -rw-r--r-- | docs/testing/user/userguide/12-nsb-overview.rst | 306 |
1 files changed, 179 insertions, 127 deletions
diff --git a/docs/testing/user/userguide/12-nsb-overview.rst b/docs/testing/user/userguide/12-nsb-overview.rst index 71a5c1130..45b087a47 100644 --- a/docs/testing/user/userguide/12-nsb-overview.rst +++ b/docs/testing/user/userguide/12-nsb-overview.rst @@ -1,77 +1,98 @@ .. This work is licensed under a Creative Commons Attribution 4.0 International .. License. .. http://creativecommons.org/licenses/by/4.0 -.. (c) OPNFV, 2016-2017 Intel Corporation. +.. (c) OPNFV, 2016-2019 Intel Corporation. + +.. Convention for heading levels in Yardstick documentation: + + ======= Heading 0 (reserved for the title in a document) + ------- Heading 1 + ^^^^^^^ Heading 2 + +++++++ Heading 3 + ''''''' Heading 4 + + Avoid deeper levels because they do not render well. =================================== Network Services Benchmarking (NSB) =================================== -Abstract -======== +.. _Yardstick: https://wiki.opnfv.org/display/yardstick +.. _`ETSI GS NFV-TST001`: http://www.etsi.org/deliver/etsi_gs/NFV-TST/001_099/001/01.01.01_60/gs_nfv-tst001v010101p.pdf -.. _Yardstick: https://wiki.opnfv.org/yardstick +Abstract +-------- This chapter provides an overview of the NSB, a contribution to OPNFV Yardstick_ from Intel. Overview -======== - -The goal of NSB is to Extend Yardstick to perform real world VNFs and NFVi -Characterization and benchmarking with repeatable and deterministic methods. - -The Network Service Benchmarking (NSB) extends the yardstick framework to do -VNF characterization and benchmarking in three different execution -environments - bare metal i.e. native Linux environment, standalone virtual -environment and managed virtualized environment (e.g. Open stack etc.). -It also brings in the capability to interact with external traffic generators -both hardware & software based for triggering and validating the traffic -according to user defined profiles. +-------- + +Network Services Benchmarking (:term:`NSB`) uses the :term:`Yardstick` +framework for performing :term:`VNF` and :term:`NFVI` characterisation in an +:term:`NFV` environment. + +For VNF characterisation, NSB will onboard a VNF, source and sink traffic to it +via traffic generators, and collect a variety of key performance indicators +(:term:`KPI`) during VNF execution. The stream of KPI data is stored in a +database, and it is visualized in a performance-visualization dashboard. + +For NFVI characterisation, a fixed test VNF, called :term:`PROX` is used. +PROX implements a suite of test cases and visualizes the output data of the +test suite. The PROX test cases implement various execution kernels found in +real-world VNFs, and the output of the test cases provides an indication of +the fitness of the infrastructure for running NFV services, in addition to +indicating potential performance optimizations for the NFVI. + +NSB extends the Yardstick framework to do VNF characterization in three +different execution environments - bare metal i.e. native Linux environment, +standalone virtual environment and managed virtualized environment (e.g. +OpenStack). It also brings in the capability to interact with external traffic +generators, both hardware and software based, for triggering and validating the +traffic according to user defined profiles. NSB extension includes: - - Generic data models of Network Services, based on ETSI spec `ETSI GS NFV-TST 001 <http://www.etsi.org/deliver/etsi_gs/NFV-TST/001_099/001/01.01.01_60/gs_nfv-tst001v010101p.pdf>`_ - - - New Standalone context for VNF testing like SRIOV, OVS, OVS-DPDK etc - - - Generic VNF configuration models and metrics implemented with Python - classes - - - Traffic generator features and traffic profiles - - - L1-L3 state-less traffic profiles +* Generic data models of Network Services, based on ETSI spec + `ETSI GS NFV-TST001`_ +* Standalone :term:`context` for VNF testing SRIOV, OVS-DPDK, etc +* Generic VNF configuration models and metrics implemented with Python + classes +* Traffic generator features and traffic profiles - - L4-L7 state-full traffic profiles + * L1-L3 stateless traffic profiles + * L4-L7 state-full traffic profiles + * Tunneling protocol/network overlay support - - Tunneling protocol / network overlay support +* Scenarios that handle NSB test cases execution - - Test case samples + * NSPerf - scenario that handles generic NSB test case execution + (setup and init tg/vnf, trigger traffic on tg, collect kpi) + * NSPerf-RFC2544 - scenario that allows repeatable triggering of traffic on + traffic generators until test case acceptance criteria is met + (for example RFC2544 binary search) - - Ping +* Test case samples - - Trex + * Ping + * Trex + * vPE, vCGNAT, vFirewall etc - ipv4 throughput, latency etc - - vPE,vCGNAT, vFirewall etc - ipv4 throughput, latency etc +* Traffic generators i.e. Trex, ab/nginx, ixia, iperf, etc +* KPIs for a given use case: - - Traffic generators like Trex, ab/nginx, ixia, iperf etc + * System agent support for collecting NFVi KPI. This includes: - - KPIs for a given use case: + * CPU statistic + * Memory BW + * OVS-DPDK Stats - - System agent support for collecting NFVi KPI. This includes: - - - CPU statistic - - - Memory BW - - - OVS-DPDK Stats - - - Network KPIs, e.g., inpackets, outpackets, thoughput, latency etc - - - VNF KPIs, e.g., packet_in, packet_drop, packet_fwd etc + * Network KPIs e.g. inpackets, outpackets, thoughput, latency + * VNF KPIs e.g. packet_in, packet_drop, packet_fwd Architecture -============ +------------ The Network Service (NS) defines a set of Virtual Network Functions (VNF) connected together using NFV infrastructure. @@ -83,124 +104,155 @@ performed network functionality. The part of the data model is a set of the configuration parameters, number of connection points used and flavor including core and memory amount. -The ETSI defines a Network Service as a set of configurable VNFs working in -some NFV Infrastructure connecting each other using Virtual Links available -through Connection Points. The ETSI MANO specification defines a set of -management entities called Network Service Descriptors (NSD) and -VNF Descriptors (VNFD) that define real Network Service. The picture below -makes an example how the real Network Operator use-case can map into ETSI -Network service definition +ETSI defines a Network Service as a set of configurable VNFs working in some +NFV Infrastructure connecting each other using Virtual Links available through +Connection Points. The ETSI MANO specification defines a set of management +entities called Network Service Descriptors (NSD) and VNF Descriptors (VNFD) +that define real Network Service. The picture below makes an example how the +real Network Operator use-case can map into ETSI Network service definition. + +Network Service framework performs the necessary test steps. It may involve: + +* Interacting with traffic generator and providing the inputs on traffic + type / packet structure to generate the required traffic as per the + test case. Traffic profiles will be used for this. +* Executing the commands required for the test procedure and analyses the + command output for confirming whether the command got executed correctly + or not e.g. as per the test case, run the traffic for the given + time period and wait for the necessary time delay. +* Verify the test result. +* Validate the traffic flow from SUT. +* Fetch the data from SUT and verify the value as per the test case. +* Upload the logs from SUT onto the Test Harness server +* Retrieve the KPI's provided by particular VNF -Network Service framework performs the necessary test steps. It may involve - - - Interacting with traffic generator and providing the inputs on traffic - type / packet structure to generate the required traffic as per the - test case. Traffic profiles will be used for this. +Components of Network Service +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - - Executing the commands required for the test procedure and analyses the - command output for confirming whether the command got executed correctly - or not. E.g. As per the test case, run the traffic for the given - time period / wait for the necessary time delay +.. TODO: provide a list of components in this section and describe them in + later sub-sections - - Verify the test result. +.. Components are the methodology, TGs, framework extensions, KPI collection, + Testcases, SampleVNFs +.. Framework extentions include: VNF models, NSPerf Scenario, contexts - - Validate the traffic flow from SUT +* *Models for Network Service benchmarking*: The Network Service benchmarking + requires the proper modelling approach. The NSB provides models using Python + files and defining of NSDs and VNFDs. - - Fetch the table / data from SUT and verify the value as per the test case +The benchmark control application being a part of OPNFV Yardstick can call +that Python models to instantiate and configure the VNFs. Depending on +infrastructure type (bare-metal or fully virtualized) that calls could be +made directly or using MANO system. - - Upload the logs from SUT onto the Test Harness server +* *Traffic generators in NSB*: Any benchmark application requires a set of + traffic generator and traffic profiles defining the method in which traffic + is generated. - - Read the KPI's provided by particular VNF +The Network Service benchmarking model extends the Network Service +definition with a set of Traffic Generators (TG) that are treated +same way as other VNFs being a part of benchmarked network service. +Same as other VNFs the traffic generator are instantiated and terminated. -Components of Network Service ------------------------------ +Every traffic generator has own configuration defined as a traffic profile +and a set of KPIs supported. The python models for TG is extended by +specific calls to listen and generate traffic. - * *Models for Network Service benchmarking*: The Network Service benchmarking - requires the proper modelling approach. The NSB provides models using Python - files and defining of NSDs and VNFDs. +* *The stateless TREX traffic generator*: The main traffic generator used as + Network Service stimulus is open source TREX tool. - The benchmark control application being a part of OPNFV yardstick can call - that python models to instantiate and configure the VNFs. Depending on - infrastructure type (bare-metal or fully virtualized) that calls could be - made directly or using MANO system. +The TREX tool can generate any kind of stateless traffic. - * *Traffic generators in NSB*: Any benchmark application requires a set of - traffic generator and traffic profiles defining the method in which traffic - is generated. - - The Network Service benchmarking model extends the Network Service - definition with a set of Traffic Generators (TG) that are treated - same way as other VNFs being a part of benchmarked network service. - Same as other VNFs the traffic generator are instantiated and terminated. +.. code-block:: console - Every traffic generator has own configuration defined as a traffic profile - and a set of KPIs supported. The python models for TG is extended by - specific calls to listen and generate traffic. + +--------+ +-------+ +--------+ + | | | | | | + | Trex | ---> | VNF | ---> | Trex | + | | | | | | + +--------+ +-------+ +--------+ - * *The stateless TREX traffic generator*: The main traffic generator used as - Network Service stimulus is open source TREX tool. +Supported testcases scenarios: - The TREX tool can generate any kind of stateless traffic. +* Correlated UDP traffic using TREX traffic generator and replay VNF. - .. code-block:: console + * Using different IMIX configuration like pure voice, pure video traffic etc + * Using different number IP flows e.g. 1, 1K, 16K, 64K, 256K, 1M flows + * Using different number of rules configured e.g. 1, 1K, 10K rules - +--------+ +-------+ +--------+ - | | | | | | - | Trex | ---> | VNF | ---> | Trex | - | | | | | | - +--------+ +-------+ +--------+ +For UDP correlated traffic following Key Performance Indicators are collected +for every combination of test case parameters: - Supported testcases scenarios: +* RFC2544 throughput for various loss rate defined (1% is a default) - - Correlated UDP traffic using TREX traffic generator and replay VNF. +KPI Collection +^^^^^^^^^^^^^^ - - using different IMIX configuration like pure voice, pure video traffic etc +KPI collection is the process of sampling KPIs at multiple intervals to allow +for investigation into anomalies during runtime. Some KPI intervals are +adjustable. KPIs are collected from traffic generators and NFVI for the SUT. +There is already some reporting in NSB available, but NSB collects all KPIs for +analytics to process. - - using different number IP flows like 1 flow, 1K, 16K, 64K, 256K, 1M flows +Below is an example list of basic KPIs: - - Using different number of rules configured like 1 rule, 1K, 10K rules +* Throughput +* Latency +* Packet delay variation +* Maximum establishment rate +* Maximum tear-down rate +* Maximum simultaneous number of sessions - For UDP correlated traffic following Key Performance Indicators are collected - for every combination of test case parameters: +Of course, there can be many other KPIs that will be relevant for a specific +NFVI, but in most cases these KPIs are enough to give you a basic picture of +the SUT. NSB also uses :term:`collectd` in order to collect the KPIs. Currently +the following collectd plug-ins are enabled for NSB testcases: - - RFC2544 throughput for various loss rate defined (1% is a default) +* Libvirt +* Interface stats +* OvS events +* vSwitch stats +* Huge Pages +* RAM +* CPU usage +* Intel® PMU +* Intel® RDT Graphical Overview -================== +------------------ -NSB Testing with yardstick framework facilitate performance testing of various +NSB Testing with Yardstick framework facilitate performance testing of various VNFs provided. .. code-block:: console +-----------+ - | | +-----------+ - | vPE | ->|TGen Port 0| - | TestCase | | +-----------+ - | | | - +-----------+ +------------------+ +-------+ | - | | -- API --> | VNF | <---> - +-----------+ | Yardstick | +-------+ | - | Test Case | --> | NSB Testing | | - +-----------+ | | | - | | | | - | +------------------+ | - +-----------+ | +-----------+ - | Traffic | ->|TGen Port 1| - | patterns | +-----------+ + | | +-------------+ + | vPE | -->| TGen Port 0 | + | TestCase | | +-------------+ + | | | + +-----------+ +---------------+ +-------+ | + | | ---> | VNF | <---> + +-----------+ | Yardstick | +-------+ | + | Test Case | --> | NSB Testing | | + +-----------+ | | | + | | | | + | +---------------+ | + +-----------+ | +-------------+ + | Traffic | -->| TGen Port 1 | + | patterns | +-------------+ +-----------+ Figure 1: Network Service - 2 server configuration -VNFs supported for chracterization: ------------------------------------ +VNFs supported for chracterization +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 1. CGNAPT - Carrier Grade Network Address and port Translation 2. vFW - Virtual Firewall 3. vACL - Access Control List -4. Prox - Packet pROcessing eXecution engine: - - VNF can act as Drop, Basic Forwarding (no touch), - L2 Forwarding (change MAC), GRE encap/decap, Load balance based on - packet fields, Symmetric load balancing - - QinQ encap/decap IPv4/IPv6, ARP, QoS, Routing, Unmpls, Policing, ACL +4. PROX - Packet pROcessing eXecution engine: + * VNF can act as Drop, Basic Forwarding (no touch), + L2 Forwarding (change MAC), GRE encap/decap, Load balance based on + packet fields, Symmetric load balancing + * QinQ encap/decap IPv4/IPv6, ARP, QoS, Routing, Unmpls, Policing, ACL 5. UDP_Replay |