diff options
Diffstat (limited to 'docs')
15 files changed, 505 insertions, 326 deletions
diff --git a/docs/scenarios/os-nosdn-fdio-ha/FDS-nosdn-overview.png b/docs/scenarios/os-nosdn-fdio-ha/FDS-nosdn-overview.png Binary files differnew file mode 100755 index 0000000..0692374 --- /dev/null +++ b/docs/scenarios/os-nosdn-fdio-ha/FDS-nosdn-overview.png diff --git a/docs/scenarios/os-nosdn-fdio-ha/index.rst b/docs/scenarios/os-nosdn-fdio-ha/index.rst new file mode 100644 index 0000000..f944346 --- /dev/null +++ b/docs/scenarios/os-nosdn-fdio-ha/index.rst @@ -0,0 +1,18 @@ +.. OPNFV - Open Platform for Network Function Virtualization +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + + +********************************************************************** +Fast Data Stacks Scenario: os-nosdn-fdio-ha Overview and Description +********************************************************************** + +Scenario: "OpenStack - FD.io" (os-nosdn-fdio-ha) +is a scenario developed as part of the FastDataStacks +OPNFV project. + +.. toctree:: + :numbered: + :maxdepth: 2 + + scenario.description.rst diff --git a/docs/scenarios/os-nosdn-fdio-ha/scenario.description.rst b/docs/scenarios/os-nosdn-fdio-ha/scenario.description.rst new file mode 100644 index 0000000..86eadb2 --- /dev/null +++ b/docs/scenarios/os-nosdn-fdio-ha/scenario.description.rst @@ -0,0 +1,173 @@ +.. OPNFV - Open Platform for Network Function Virtualization +.. This work is licensed under a Creative Commons Attribution 4.0 +.. International License. +.. http://creativecommons.org/licenses/by/4.0 + +Scenario: "OpenStack - FD.io" +============================= + +Scenario: os-nosdn-fdio-ha + +"os-nosdn-fdio-ha" is a scenario developed as part of the FastDataStacks +OPNFV project. The main components of the "os-nosdn-fdio-ha" scenario +are: + + - APEX (TripleO) installer (please also see APEX installer documentation) + - Openstack (in HA configuration) + - FD.io/VPP virtual forwarder for tenant networking + - etcd, which is the VPP ML2 mechanism driver's distributed key-value store, in clustered mode + +Introduction +============ + +NFV and virtualized high performance applications, such as video processing, +require a "fast data stack" solution that provides both carrier grade +forwarding performance, scalability and open extensibility. + +A key component of any NFV solution is the virtual forwarder, which needs to be +a feature rich, high performance, highly scale virtual switch-router. It needs +to leverage hardware accelerators when available and run in user space. In +addition, it should be modular and easily extensible. The Vector Packet +Processor (VPP) supplied by the FD.io project meets these needs, in that +it offers a highly scalable, high performance and easily extensible +software forwarder that entirely runs in user space. + +The "Openstack - FD.io/VPP" scenario provides the capability to realize a set +of use-cases relevant to the deployment of NFV nodes instantiated by means of +an Openstack orchestration system on FD.io/VPP enabled compute nodes. + +A deployment of the "os-nosdn-fdio-ha" scenario consists of 6 or more +servers: + + * 1 Jumphost hosting the APEX installer - running the Undercloud + * 3 Controlhosts, which run the Overcloud and Openstack services as well as the VPP ML2 etcd cluster + * 2 or more Computehosts + +.. image:: FDS-nosdn-overview.png + +Tenant networking leverages FD.io/VPP. Open VSwitch (OVS) is used for all other +connectivity, in particular the connectivity to public networking / the +Internet (i.e. br-ext) is performed via OVS as in any standard OpenStack +deployment. A VPP management agent is used to setup and manage layer 2 +networking for the scenario. Neutron ML2 plugin is configured to use +the VPP ML2 networking mechanism driver. Tenant networking can either leverage +VLANs or plain interfaces (flat networks). Layer 3 connectivity for a tenant +network is provided centrally via qrouter on the control node. As in a +standard OpenStack deployment, the Layer3 agent configures the qrouter and +associated rulesets for security (security groups) and NAT (floating IPs). +Public IP network connectivity for a tenant network is provided by +interconnecting the VPP-based bridge domain representing the tenant network to +qrouter using a tap interface. + +The setup is depicted below: + +Features of the scenario +------------------------ + +Main features of the "os-nosdn-fdio-ha" scenario: + + * Automated installation using the APEX installer + * Fast and scalable tenant networking using FD.io/VPP as forwarder + * Layer 2 networking using VLANs, managed and controlled + through the VPP ML2 plugin + * Layer 3 connectivitiy for tenant networks supplied centrally + on the Control node through standard OpenStack mechanisms. + All layer 3 features apply, including floating IPs (i.e. NAT) + and security groups + * DHCP server for tenant instances provided using the standard + OpenStack dnsmasq server + * OpenStack high availability + * etcd (VPP ML2 mechanism driver's distributed key-value store) high availability + +Networking in this scenario using VPP +------------------------------------- + +The os-nosdn-fdio-ha scenario combines components from two key open +source projects: OpenStack and Fast Data (FD.io). In order to make Fast Data +(FD.io) networking available to this scenario, an ML2 mechanism driver and a +light-weight control plane agent for VPP forwarder has been created. For +details see also https://git.openstack.org/cgit/openstack/networking-vpp/ + +Networking-vpp provides a Neutron ML2 mechanism driver to bring the advantages +of VPP to OpenStack deployments.It uses an etcd cluster on the control node to +keep track of the compute nodes, agent state and port bindings/unbindings. + +It's been written to be as simple and readable as possible, which means it's +naive; the aim was not to write the most efficient mechanism driver ever from +right out of the gate, but to write something simple and understandable and see +how well it works and what needs to be changed. + +As a general rule, everything was implemented in the simplest way, for two +reasons: one is that one sees working results the quickes, and the other is +that it's much easier to replace a simple system with a more complex one than +it is to change a complex one. The current design will change, but the one +that's there at the moment is small and easy to read, even if it makes you pull +faces when you read it. + +Scenario Configuration +====================== + +To enable the "os-nosdn-fdio-ha" scenario check the appropriate settings +in the APEX configuration files. Those are typically found in /etc/opnfv-apex. + +Use the file "os-nosdn-fdio-ha.yaml":: + + global_params: + ha_enabled: true + + deploy_options: + sdn_controller: false + sdn_l3: false + tacker: true + congress: true + sfc: false + vpn: false + vpp: true + dataplane: fdio + performance: + Controller: + vpp: + uio-driver: uio_pci_generic + Compute: + kernel: + hugepagesz: 2M + hugepages: 2048 + intel_iommu: 'on' + iommu: pt + isolcpus: 1,2 + vpp: + uio-driver: uio_pci_generic + + +Validated deployment environments +================================= + +The "os-nosdn-fdio-ha" scenario has been deployed and tested +on the following sets of hardware: + * Linux Foundation lab (Chassis: Cisco UCS-B-5108 blade server, + NICs: 8 external / 32 internal 10GE ports, + RAM: 32G (4 x 8GB DDR4-2133-MHz RDIMM/PC4-17000/single rank/x4/1.2v), + CPU: 3.50 GHz E5-2637 v3/135W 4C/15MB Cache/DDR4 2133MHz + Disk: 1.2 TB 6G SAS 10K rpm SFF HDD) see also + https://wiki.opnfv.org/display/pharos/Lflab+Hosting + * Cisco internal development labs (UCS-B and UCS-C) + + +Limitations, Issues and Workarounds +=================================== + +For specific information on limitations and issues, please refer to the APEX +installer release notes. + +References +========== + + + * FastDataStacks OPNFV project wiki: https://wiki.opnfv.org/display/fds + * Fast Data (FD.io): https://fd.io/ + * FD.io Vector Packet Processor (VPP): https://wiki.fd.io/view/VPP + * ML2 VPP mechanisms driver: https://git.openstack.org/cgit/openstack/networking-vpp/ + * OPNFV Danube release - more information: http://www.opnfv.org/danube + * Networking-vpp launchpad (ticket tracker) https://launchpad.net/networking-vpp + * Networking-vpp Wiki: https://wiki.openstack.org/wiki/Networking-vpp/ + * APEX (TripleO based) installer: https://wiki.opnfv.org/display/apex/Apex diff --git a/docs/scenarios/os-odl_l2-fdio-ha/FDS-basic-components.jpg b/docs/scenarios/os-odl_l2-fdio-ha/FDS-basic-components.jpg Binary files differnew file mode 100755 index 0000000..e92851f --- /dev/null +++ b/docs/scenarios/os-odl_l2-fdio-ha/FDS-basic-components.jpg diff --git a/docs/scenarios/os-odl_l2-fdio-ha/FDS-odl_l2-ha-overview.png b/docs/scenarios/os-odl_l2-fdio-ha/FDS-odl_l2-ha-overview.png Binary files differnew file mode 100755 index 0000000..78526da --- /dev/null +++ b/docs/scenarios/os-odl_l2-fdio-ha/FDS-odl_l2-ha-overview.png diff --git a/docs/scenarios/os-odl_l2-fdio-ha/FDS-simple-callflow.png b/docs/scenarios/os-odl_l2-fdio-ha/FDS-simple-callflow.png Binary files differnew file mode 100755 index 0000000..04546aa --- /dev/null +++ b/docs/scenarios/os-odl_l2-fdio-ha/FDS-simple-callflow.png diff --git a/docs/scenarios/os-odl_l2-fdio-ha/scenario.description.rst b/docs/scenarios/os-odl_l2-fdio-ha/scenario.description.rst index d8787c6..a81e8ed 100755 --- a/docs/scenarios/os-odl_l2-fdio-ha/scenario.description.rst +++ b/docs/scenarios/os-odl_l2-fdio-ha/scenario.description.rst @@ -13,7 +13,7 @@ FastDataStacks OPNFV project. The main components of the - APEX (TripleO) installer (please also see APEX installer documentation) - Openstack (in HA configuration) - - OpenDaylight controller (non-clustered) controlling layer 2 networking + - OpenDaylight controller in clustered mode controlling layer 2 networking - FD.io/VPP virtual forwarder for tenant networking Introduction @@ -39,8 +39,8 @@ NFV infrastructure are reflect different business In order to meet the desired qualities of an NFV infrastructure, the -following components were chosen for the "Openstack - OpenDaylight - - FD.io/VPP" scenario: +following components were chosen for the "Openstack - OpenDaylight - FD.io" +scenario: * FD.io Vector Packet Processor (VPP) - a highly scalable, high performance, extensible virtual forwarder * OpenDaylight Controller - an extensible controller platform which @@ -48,9 +48,9 @@ following components were chosen for the "Openstack - OpenDaylight constructs, supports a diverse set of network devices (virtual and physical) via the "group based policy (GBP)" component, and can be clustered to achieve a highly available - deployment. + deployment - as done in this scenario. -The "Openstack - OpenDaylight - FD.io/VPP" scenario provides the capability to +The "Openstack - OpenDaylight - FD.io" scenario provides the capability to realize a set of use-cases relevant to the deployment of NFV nodes instantiated by means of an Openstack orchestration system on FD.io/VPP enabled compute nodes. The role of the Opendaylight network controller in this integration is @@ -70,11 +70,12 @@ A deployment of the "apex-os-odl_l2-fdio-ha" scenario consists of 4 or more servers: * 1 Jumphost hosting the APEX installer - running the Undercloud - * 3 Controlhosts, which runs the Overcloud as well as OpenDaylight - as a network controller (OpenDaylight only runs on one Controlhost) + * 3 Controlhosts, which run the Overcloud as well as OpenDaylight + as a network controller. OpenDaylight is deployed in clustered + mode and runs on all 3 control nodes. * 2 or more Computehosts -.. image:: FDS-odl_l2-overview.png +.. image:: FDS-odl_l2-ha-overview.png Tenant networking leverages FD.io/VPP. Open VSwitch (OVS) is used for all other connectivity, in particular the connectivity to public networking / the @@ -113,6 +114,7 @@ Main features of the "apex-os-odl_l2-fdio-ha" scenario: All layer 3 features apply, including floating IPs (i.e. NAT) and security groups. * Manual and automatic (via DHCP) addressing on tenant networks + * OpenDaylight controller high availability (clustering) * OpenStack high availability Scenario components and composition @@ -128,88 +130,39 @@ Vector Packet Processor (VPP). Here's a more detailed list of the individual software components involved: -**Openstack Neutron ML2 ODL Plugin**: Handles Neutron data base synchronization -and interaction with the southbound Openstack controller using HTTP. - -**OpenDaylight Neutron Nothbound & Neutron MD-SAL Entry Store**: Presents a -Neutron (v2) extended HTTP API servlet for interaction with Openstack Neutron. -It validates and stores the received Neutron data in the MD-SAL data store -against the Neutron yang model driven. - -**OpenDaylight Neutron Mapper**: The Neutron Mapper listens to Neutron data -change events and is responsible for using Neutron data in creating Group Based -Policy Data objects, e.g. GBP End-Points, Flood-Domains. A GBP End Point -represents a specific NFV/VM port and its identity as derived from a Neutron -Port. The mapped data is stored using the GBP End Point yang model and an -association between the GBP End-Point and its Neutron object is maintained in -the Neutron-GBP map. - -**OpenDaylight Group Based Policy (GBP) Entities store**: Stores for the GBP -data artifacts against the GBP YANG schemas. - -**Neutron Group Based Policy Map store**: Stores the bi-lateral relation -between an End-Point and its corresponding Neutron object. Neutron-GBP map; -keyed by Neutron object type, port, and Neutron UUID, gives the GBP End-Point, -Flood domain respectively. GBP-Neutron map keyed by GBP object type, end-point. - -**Neutron VPP Renderer Mapper**: The Neutron VPP Renderer Mapper listens to -Neutron Store data change events, as well as being able to access directly the -store, and is responsible for converting Neutron data specifically required to -render a VPP node configuration with a given End Point, e.g. the virtual host -interface name assigned to a vhostuser socket.. The mapped data is stored in -the VPP info data store. - -**VPP Info Store**: Stores VPP specific information regarding End-Points, Flood -domains with VLAN, etc. - -**GBP Renderer Manager**: The GBP Renderer Manager is the central point for -dispatching of data to specific device renderers. It uses the information -derived from the GBP end-point and its topology entries to dispatch the task of -configuration to a specific device renderer by writing a renderer policy -configuration into the registered renderer's policy store. The renderer manager -also monitors, by being a data change listener on the VPP Renderer Policy -States, for any errors in the application of a rendered configuration. - -**Renderer Policy Config Store**: The store's schema serves as the API between -the Renderer Manager and specific Renderers like the VPP Renderer. The store -uses a a YANG modeled schema to represent all end-point and associated GBP -policy data. - -**Topology Entries Store**: The yang model based MD-SAL topology store serves -two fundamental roles: 1. It maintains a topological representation of the GBP -End Points, in the context of customer networks. 2. It maintains an association -of each (VPP) compute node's physical interfaces to their neutron provider -network (e.g. The association between an ethernet interface and a Neutron -provider network). - -**VPP Renderer**: The VPP Renderer registers an instance for VPP nodes with the -Renderer Manager by means of inserting operational data into the Renderer -Policy config store. It acts as a listener on the Renderer Policy consumes via -the GBP Policy API data + the specific VPP End Point data, to drive the -configuration of VPP devices using NETCONF Services. -More specifically, the renderer generates: - - * vhost user port configuration that corresponds to the VM port configuration - * VPP bridge instances corresponding to the GBP flood domain - * port or traffic filtering configuration, in accordance with the GBP policy. - -The VPP Renderer also interacts with the Virtual Bridge Domain Service, by -means of the VBD store, in order to establish connectivity between VPP nodes in -a bridge domain. For this it uses the VPP device name, and the flood domain -data derived from the VPP Info and End-Point data respectively. For the -executed configuration operations it updates state in the Renderer policy state -store. - -**Virtual Bridge Domain (VBD) Store and Manager**: The virtual bridge domain -manager is responsible for configuring the VxLAN overlay tunnel infrastructure -to arrive at a desired bridged topology between multiple (VPP) compute nodes. -VDB configures VXLAN tunnels always into a full-mesh with split-horizon group -forwarding applied on any domain facing tunnel interface (i.e. forwarding -behavior will be that used for VPLS). - -**NETCONF Mount Point Service & Connector**: Collectively referred to as -Netconf Services, provide the NETCONF interface for accessing VPP configuration -and operational data stores that are represented as NETCONF mounts. +**Openstack Neutron ML2 OpenDaylight Plugin**: Handles Neutron data base +synchronization and interaction with the southbound controller using a REST +interface. + +**ODL GBP Neutron Mapper**: Maps neutron elements like networks, subnets, +security groups, etc. to GBP entities: Creates policy and configuration for +tenants (endpoints, resolved policies, forwarding rules). + +**ODL GBP Neutron VPP Mapper**: Maps Neutron ports to VPP endpoints in GBP. + +**ODL GBP Location Manager**: Provides real location for endpoints (i.e. Which +physical node an endpoint is connected to). + +**GBP Renderer Manager**: Creates configuration for Renderers (like e.g. +VPP-Renderer or OVS-Renderer). The GBP Renderer Manager is the central point +for dispatching of data to specific device renderers. It uses the information +derived from the GBP end-point and its topology entries to dispatch the task +of configuration to a specific device renderer by writing a renderer policy +configuration into the registered renderer's policy store. The renderer +manager also monitors, by being a data change listener on the VPP Renderer +Policy States, for any errors in the application of a rendered configuration. + +**GBP VPP Renderer Interface Manager**: Listens to VPP endpoints in the +Config DataStore and configures associated interfaces on VPP via HoneyComb. + +**GBP VPP Renderer Renderer Policy Manager**: Manages the creation of +bridge domains using VBD and assigns interfaces to bridge domains. + +**Virtual Bridge Domain Manager (VBD)**: Creates bridge domains (i.e. in case +of VXLAN creates full mesh of VXLAN tunnels, configures split horizon on +tunnel endpoints etc.). VDB configures VXLAN tunnels always into a full-mesh +with split-horizon group forwarding applied on any domain facing tunnel +interface (i.e. forwarding behavior will be that used for VPLS). **Virtual Packet Processor (VPP) and Honeycomb server**: The VPP is the accelerated data plane forwarding engine relying on vhost user interfaces @@ -217,20 +170,45 @@ towards Virtual Machines created by the Nova Agent. The Honeycomb NETCONF configuration server is responsible for driving the configuration of the VPP, and collecting the operational data. -**Rendered Policy State Store**: Stores data regarding the execution of -operations performed by a given renderer. - **Nova Agent**: The Nova Agent, a sub-component of the overall Openstack architecture, is responsible for interacting with the compute node's host Libvirt API to drive the life-cycle of Virtual Machines. It, along with the compute node software, are assumed to be capable of supporting vhost user interfaces. -The picture below show a basic end to end call flow for creating a Neutron -vhostuser port on VPP using a GBP renderer. It showcases how the different -component described above interact. +The picture below shows the key components. + +.. image:: FDS-basic-components.jpg + +To provide a better understanding how the above mentioned components interact +with each other, the following diagram shows how the example of creating a +vhost-user port on VPP through Openstack Neutron: + +To create or update a port, Neutron will send a request to ODL Neutron +Northbound which contains the UUID, along with the host-id as "vpp" and +vif-type as "vhost-user". The GBP Neutron mapper turns the "Neutron speak" of +"ports" into the generic connectivity model that GroupBasedPolicy uses. +Neutron "ports" become generic "GBP Endpoints" which can be consumed by the +GBP Renderer Manager. The GBP Renderer Manager resolves the policy for the +endpoint, i.e. it determines which communication relationships apply to the +specific endpoint, and hands the resolution to a device specific renderer, +which is the VPP renderer in the given case here. VPP renderer turns the +generic policy into VPP specific configuration. Note that in case the policy +would need to be applied to a different device, e.g. an OpenVSwitch (OVS), +then an "OVS Renderer" would be used. VPP Renderer and the topology manager +("Virtual Bridge Domain" manager - i.e. VBD) cooperate to create the actual +network configuration. VPP Renderer configures the interfaces to the virtual +machines (VM), i.e. the vhost-user interface in the given case here and +attaches them to a bridge domain on VPP. VBD handles the setup of connectivity +between bridge domains on individual VPPs, i.e. it maintains the VXLAN tunnels +in the given case here. Both VPP Renderer as well as VBD communicate with the +device through Netconf/YANG. All compute and control nodes run an instance of +VPP and the VPP-configuration agent "Honeycomb". Honeycomb serves as a +Netconf/YANG server, receives the configuration commands from VBD and VPP +Renderer and drives VPP configuration using VPP's local Java APIs. + +.. image:: FDS-simple-callflow.png -.. image:: FDS-basic-callflow.jpg Scenario Configuration ====================== @@ -240,7 +218,11 @@ settings in the APEX configuration files. Those are typically found in /etc/opnfv-apex. File "deploy_settings.yaml" choose opendaylight as controller with version -"boron" and enable vpp as forwarder. "hugepages" need to set to a +"carbon" and enable vpp as forwarder. Also make sure that you set +"ha_enabled" to "true" in the global_params section. "ha_enabled" is the +only real difference from a configuration file perspective between the +scenario with high availability when compared to the ODL-L2 scenario +without high-availability support. "hugepages" need to set to a sufficiently large value for VPP to work. The default value for VPP is 1024, but this only allows for a few VMs to be started. If feasible, choose a significantly larger number on the compute nodes:: @@ -251,7 +233,7 @@ choose a significantly larger number on the compute nodes:: deploy_options: sdn_controller: opendaylight sdn_l3: false - odl_version: boron + odl_version: carbon tacker: true congress: true sfc: false @@ -265,12 +247,22 @@ choose a significantly larger number on the compute nodes:: hugepagesz: 2M intel_iommu: 'on' iommu: pt + isolcpus: 1,2 + vpp: + main-core: 1 + corelist-workers: 2 + uio-driver: uio_pci_generic Compute: kernel: hugepagesz: 2M hugepages: 2048 intel_iommu: 'on' iommu: pt + isolcpus: 1,2 + vpp: + main-core: 1 + corelist-workers: 2 + uio-driver: uio_pci_generic Validated deployment environments @@ -290,9 +282,10 @@ on the following sets of hardware: Limitations, Issues and Workarounds =================================== -There are no known issues. Note that only OpenStack is deployed in -HA mode. OpenDaylight clustering is expected to be added in a future -revision of this scenario. +For specific information on limitations and issues, please refer to the APEX +installer release notes. Note that this high availability scenario +deploys OpenStack in HA mode *and* OpenDaylight in cluster mode. + References ========== @@ -302,5 +295,5 @@ References * Fast Data (FD.io): https://fd.io/ * FD.io Vector Packet Processor (VPP): https://wiki.fd.io/view/VPP * OpenDaylight Controller: https://www.opendaylight.org/ - * OPNFV Colorado release - more information: http://www.opnfv.org/colorado + * OPNFV Danube release - more information: http://www.opnfv.org/danube diff --git a/docs/scenarios/os-odl_l2-fdio-noha/FDS-basic-components.jpg b/docs/scenarios/os-odl_l2-fdio-noha/FDS-basic-components.jpg Binary files differnew file mode 100755 index 0000000..e92851f --- /dev/null +++ b/docs/scenarios/os-odl_l2-fdio-noha/FDS-basic-components.jpg diff --git a/docs/scenarios/os-odl_l2-fdio-noha/FDS-simple-callflow.png b/docs/scenarios/os-odl_l2-fdio-noha/FDS-simple-callflow.png Binary files differnew file mode 100755 index 0000000..04546aa --- /dev/null +++ b/docs/scenarios/os-odl_l2-fdio-noha/FDS-simple-callflow.png diff --git a/docs/scenarios/os-odl_l2-fdio-noha/scenario.description.rst b/docs/scenarios/os-odl_l2-fdio-noha/scenario.description.rst index 076e37c..b392070 100755 --- a/docs/scenarios/os-odl_l2-fdio-noha/scenario.description.rst +++ b/docs/scenarios/os-odl_l2-fdio-noha/scenario.description.rst @@ -39,8 +39,8 @@ NFV infrastructure are reflect different business In order to meet the desired qualities of an NFV infrastructure, the -following components were chosen for the "Openstack - OpenDaylight - - FD.io/VPP" scenario: +following components were chosen for the "Openstack - OpenDaylight - FD.io" +scenario: * FD.io Vector Packet Processor (VPP) - a highly scalable, high performance, extensible virtual forwarder * OpenDaylight Controller - an extensible controller platform which @@ -50,7 +50,7 @@ following components were chosen for the "Openstack - OpenDaylight component, and can be clustered to achieve a highly available deployment. -The "Openstack - OpenDaylight - FD.io/VPP" scenario provides the capability to +The "Openstack - OpenDaylight - FD.io" scenario provides the capability to realize a set of use-cases relevant to the deployment of NFV nodes instantiated by means of an Openstack orchestration system on FD.io/VPP enabled compute nodes. The role of the Opendaylight network controller in this integration is @@ -76,8 +76,8 @@ servers: .. image:: FDS-odl_l2-overview.png -Tenant networking leverages FD.io/VPP. Open VSwitch (OVS) is used for all other -connectivity, in particular the connectivity to public networking / the +Tenant networking leverages FD.io/VPP. Open VSwitch (OVS) is used for all +other connectivity, in particular the connectivity to public networking / the Internet (i.e. br-ext) is performed via OVS as in any standard OpenStack deployment. The OpenDaylight network controller is used to setup and manage layer 2 networking for the scenario. Tenant networking can either leverage @@ -113,94 +113,45 @@ The apex-os-odl_l2-fdio-noha scenario combines components from three key open source projects: OpenStack, OpenDaylight, and Fast Data (FD.io). The key components that realize the apex-os-odl_l2-fdio-noha scenario and which differ from a regular, OVS-based scenario, are the OpenStack ML2 OpenDaylight plugin, -OpenDaylight Neutron Northbound, OpenDaylight Group Based Policy, OpenDaylight -Virtual Bridge Domain Manager, FD.io Honeycomb management agent and FD.io -Vector Packet Processor (VPP). +OpenDaylight Neutron Northbound, OpenDaylight (ODL) Group Based Policy (GBP), +OpenDaylight Virtual Bridge Domain Manager (VBD), FD.io Honeycomb management +agent and FD.io Vector Packet Processor (VPP). Here's a more detailed list of the individual software components involved: -**Openstack Neutron ML2 ODL Plugin**: Handles Neutron data base synchronization -and interaction with the southbound Openstack controller using HTTP. - -**OpenDaylight Neutron Nothbound & Neutron MD-SAL Entry Store**: Presents a -Neutron (v2) extended HTTP API servlet for interaction with Openstack Neutron. -It validates and stores the received Neutron data in the MD-SAL data store -against the Neutron yang model driven. - -**OpenDaylight Neutron Mapper**: The Neutron Mapper listens to Neutron data -change events and is responsible for using Neutron data in creating Group Based -Policy Data objects, e.g. GBP End-Points, Flood-Domains. A GBP End Point -represents a specific NFV/VM port and its identity as derived from a Neutron -Port. The mapped data is stored using the GBP End Point yang model and an -association between the GBP End-Point and its Neutron object is maintained in -the Neutron-GBP map. - -**OpenDaylight Group Based Policy (GBP) Entities store**: Stores for the GBP -data artifacts against the GBP YANG schemas. - -**Neutron Group Based Policy Map store**: Stores the bi-lateral relation -between an End-Point and its corresponding Neutron object. Neutron-GBP map; -keyed by Neutron object type, port, and Neutron UUID, gives the GBP End-Point, -Flood domain respectively. GBP-Neutron map keyed by GBP object type, end-point. - -**Neutron VPP Renderer Mapper**: The Neutron VPP Renderer Mapper listens to -Neutron Store data change events, as well as being able to access directly the -store, and is responsible for converting Neutron data specifically required to -render a VPP node configuration with a given End Point, e.g. the virtual host -interface name assigned to a vhostuser socket.. The mapped data is stored in -the VPP info data store. - -**VPP Info Store**: Stores VPP specific information regarding End-Points, Flood -domains with VLAN, etc. - -**GBP Renderer Manager**: The GBP Renderer Manager is the central point for -dispatching of data to specific device renderers. It uses the information -derived from the GBP end-point and its topology entries to dispatch the task of -configuration to a specific device renderer by writing a renderer policy -configuration into the registered renderer's policy store. The renderer manager -also monitors, by being a data change listener on the VPP Renderer Policy -States, for any errors in the application of a rendered configuration. - -**Renderer Policy Config Store**: The store's schema serves as the API between -the Renderer Manager and specific Renderers like the VPP Renderer. The store -uses a a YANG modeled schema to represent all end-point and associated GBP -policy data. - -**Topology Entries Store**: The yang model based MD-SAL topology store serves -two fundamental roles: 1. It maintains a topological representation of the GBP -End Points, in the context of customer networks. 2. It maintains an association -of each (VPP) compute node's physical interfaces to their neutron provider -network (e.g. The association between an ethernet interface and a Neutron -provider network). - -**VPP Renderer**: The VPP Renderer registers an instance for VPP nodes with the -Renderer Manager by means of inserting operational data into the Renderer -Policy config store. It acts as a listener on the Renderer Policy consumes via -the GBP Policy API data + the specific VPP End Point data, to drive the -configuration of VPP devices using NETCONF Services. -More specifically, the renderer generates: - - * vhost user port configuration that corresponds to the VM port configuration - * VPP bridge instances corresponding to the GBP flood domain - * port or traffic filtering configuration, in accordance with the GBP policy. - -The VPP Renderer also interacts with the Virtual Bridge Domain Service, by -means of the VBD store, in order to establish connectivity between VPP nodes in -a bridge domain. For this it uses the VPP device name, and the flood domain -data derived from the VPP Info and End-Point data respectively. For the -executed configuration operations it updates state in the Renderer policy state -store. - -**Virtual Bridge Domain (VBD) Store and Manager**: The virtual bridge domain -manager is responsible for configuring the VxLAN overlay tunnel infrastructure -to arrive at a desired bridged topology between multiple (VPP) compute nodes. -VDB configures VXLAN tunnels always into a full-mesh with split-horizon group -forwarding applied on any domain facing tunnel interface (i.e. forwarding -behavior will be that used for VPLS). - -**NETCONF Mount Point Service & Connector**: Collectively referred to as -Netconf Services, provide the NETCONF interface for accessing VPP configuration -and operational data stores that are represented as NETCONF mounts. +**Openstack Neutron ML2 OpenDaylight Plugin**: Handles Neutron data base +synchronization and interaction with the southbound controller using a REST +interface. + +**ODL GBP Neutron Mapper**: Maps neutron elements like networks, subnets, +security groups, etc. to GBP entities: Creates policy and configuration for +tenants (endpoints, resolved policies, forwarding rules). + +**ODL GBP Neutron VPP Mapper**: Maps Neutron ports to VPP endpoints in GBP. + +**ODL GBP Location Manager**: Provides real location for endpoints (i.e. Which +physical node an endpoint is connected to). + +**GBP Renderer Manager**: Creates configuration for Renderers (like e.g. +VPP-Renderer or OVS-Renderer). The GBP Renderer Manager is the central point +for dispatching of data to specific device renderers. It uses the information +derived from the GBP end-point and its topology entries to dispatch the task +of configuration to a specific device renderer by writing a renderer policy +configuration into the registered renderer's policy store. The renderer +manager also monitors, by being a data change listener on the VPP Renderer +Policy States, for any errors in the application of a rendered configuration. + +**GBP VPP Renderer Interface Manager**: Listens to VPP endpoints in the +Config DataStore and configures associated interfaces on VPP via HoneyComb. + +**GBP VPP Renderer Renderer Policy Manager**: Manages the creation of +bridge domains using VBD and assigns interfaces to bridge domains. + +**Virtual Bridge Domain Manager (VBD)**: Creates bridge domains (i.e. in case +of VXLAN creates full mesh of VXLAN tunnels, configures split horizon on +tunnel endpoints etc.). VDB configures VXLAN tunnels always into a full-mesh +with split-horizon group forwarding applied on any domain facing tunnel +interface (i.e. forwarding behavior will be that used for VPLS). **Virtual Packet Processor (VPP) and Honeycomb server**: The VPP is the accelerated data plane forwarding engine relying on vhost user interfaces @@ -208,20 +159,45 @@ towards Virtual Machines created by the Nova Agent. The Honeycomb NETCONF configuration server is responsible for driving the configuration of the VPP, and collecting the operational data. -**Rendered Policy State Store**: Stores data regarding the execution of -operations performed by a given renderer. - **Nova Agent**: The Nova Agent, a sub-component of the overall Openstack architecture, is responsible for interacting with the compute node's host Libvirt API to drive the life-cycle of Virtual Machines. It, along with the compute node software, are assumed to be capable of supporting vhost user interfaces. -The picture below show a basic end to end call flow for creating a Neutron -vhostuser port on VPP using a GBP renderer. It showcases how the different -component described above interact. +The picture below shows the key components. + +.. image:: FDS-basic-components.jpg + +To provide a better understanding how the above mentioned components interact +with each other, the following diagram shows how the example of creating a +vhost-user port on VPP through Openstack Neutron: + +To create or update a port, Neutron will send a request to ODL Neutron +Northbound which contains the UUID, along with the host-id as "vpp" and +vif-type as "vhost-user". The GBP Neutron mapper turns the "Neutron speak" of +"ports" into the generic connectivity model that GroupBasedPolicy uses. +Neutron "ports" become generic "GBP Endpoints" which can be consumed by the +GBP Renderer Manager. The GBP Renderer Manager resolves the policy for the +endpoint, i.e. it determines which communication relationships apply to the +specific endpoint, and hands the resolution to a device specific renderer, +which is the VPP renderer in the given case here. VPP renderer turns the +generic policy into VPP specific configuration. Note that in case the policy +would need to be applied to a different device, e.g. an OpenVSwitch (OVS), +then an "OVS Renderer" would be used. VPP Renderer and the topology manager +("Virtual Bridge Domain" manager - i.e. VBD) cooperate to create the actual +network configuration. VPP Renderer configures the interfaces to the virtual +machines (VM), i.e. the vhost-user interface in the given case here and +attaches them to a bridge domain on VPP. VBD handles the setup of connectivity +between bridge domains on individual VPPs, i.e. it maintains the VXLAN tunnels +in the given case here. Both VPP Renderer as well as VBD communicate with the +device through Netconf/YANG. All compute and control nodes run an instance of +VPP and the VPP-configuration agent "Honeycomb". Honeycomb serves as a +Netconf/YANG server, receives the configuration commands from VBD and VPP +Renderer and drives VPP configuration using VPP's local Java APIs. + +.. image:: FDS-simple-callflow.png -.. image:: FDS-basic-callflow.jpg Scenario Configuration ====================== @@ -231,10 +207,10 @@ settings in the APEX configuration files. Those are typically found in /etc/opnfv-apex. File "deploy_settings.yaml" choose opendaylight as controller with version -"boron" and enable vpp as forwarder. "hugepages" need to set to a -sufficiently large value for VPP to work. The default value for VPP is -1024, but this only allows for a few VMs to be started. If feasible, -choose a significantly larger number on the compute nodes:: +"carbon" and enable vpp as forwarder. "hugepages" need to set to a +sufficiently large value for VPP to work. The default value for VPP is 1024, +but this only allows for a few VMs to be started. If feasible, choose a +significantly larger number on the compute nodes:: global_params: ha_enabled: false @@ -242,7 +218,7 @@ choose a significantly larger number on the compute nodes:: deploy_options: sdn_controller: opendaylight sdn_l3: false - odl_version: boron + odl_version: carbon tacker: true congress: true sfc: false @@ -256,14 +232,22 @@ choose a significantly larger number on the compute nodes:: hugepagesz: 2M intel_iommu: 'on' iommu: pt + isolcpus: 1,2 + vpp: + main-core: 1 + corelist-workers: 2 + uio-driver: uio_pci_generic Compute: - nova: - libvirtpin: 1 kernel: hugepagesz: 2M hugepages: 2048 intel_iommu: 'on' iommu: pt + isolcpus: 1,2 + vpp: + main-core: 1 + corelist-workers: 2 + uio-driver: uio_pci_generic Validated deployment environments @@ -283,7 +267,7 @@ on the following sets of hardware: Limitations, Issues and Workarounds =================================== -There are no known issues. +For specific information on limitations and issues, please refer to the APEX installer release notes. References ========== @@ -293,5 +277,5 @@ References * Fast Data (FD.io): https://fd.io/ * FD.io Vector Packet Processor (VPP): https://wiki.fd.io/view/VPP * OpenDaylight Controller: https://www.opendaylight.org/ - * OPNFV Colorado release - more information: http://www.opnfv.org/colorado + * OPNFV Carbon release - more information: http://www.opnfv.org/carbon diff --git a/docs/scenarios/os-odl_l3-fdio-noha/FDS-L3-noha-sample-setup.png b/docs/scenarios/os-odl_l3-fdio-noha/FDS-L3-noha-sample-setup.png Binary files differnew file mode 100755 index 0000000..27c8335 --- /dev/null +++ b/docs/scenarios/os-odl_l3-fdio-noha/FDS-L3-noha-sample-setup.png diff --git a/docs/scenarios/os-odl_l3-fdio-noha/FDS-basic-components.jpg b/docs/scenarios/os-odl_l3-fdio-noha/FDS-basic-components.jpg Binary files differnew file mode 100755 index 0000000..e92851f --- /dev/null +++ b/docs/scenarios/os-odl_l3-fdio-noha/FDS-basic-components.jpg diff --git a/docs/scenarios/os-odl_l3-fdio-noha/FDS-odl_l3-noha-overview.png b/docs/scenarios/os-odl_l3-fdio-noha/FDS-odl_l3-noha-overview.png Binary files differnew file mode 100755 index 0000000..1193ea4 --- /dev/null +++ b/docs/scenarios/os-odl_l3-fdio-noha/FDS-odl_l3-noha-overview.png diff --git a/docs/scenarios/os-odl_l3-fdio-noha/FDS-simple-callflow.png b/docs/scenarios/os-odl_l3-fdio-noha/FDS-simple-callflow.png Binary files differnew file mode 100755 index 0000000..04546aa --- /dev/null +++ b/docs/scenarios/os-odl_l3-fdio-noha/FDS-simple-callflow.png diff --git a/docs/scenarios/os-odl_l3-fdio-noha/scenario.description.rst b/docs/scenarios/os-odl_l3-fdio-noha/scenario.description.rst index bf32eb6..6bbf12b 100755 --- a/docs/scenarios/os-odl_l3-fdio-noha/scenario.description.rst +++ b/docs/scenarios/os-odl_l3-fdio-noha/scenario.description.rst @@ -40,8 +40,8 @@ NFV infrastructure are reflect different business In order to meet the desired qualities of an NFV infrastructure, the -following components were chosen for the "Openstack - OpenDaylight - - FD.io/VPP" scenario: +following components were chosen for the "Openstack - OpenDaylight - FD.io" +scenario: * FD.io Vector Packet Processor (VPP) - a highly scalable, high performance, extensible virtual forwarder * OpenDaylight Controller - an extensible controller platform which @@ -52,7 +52,7 @@ following components were chosen for the "Openstack - OpenDaylight deployment. -The "Openstack - OpenDaylight - FD.io/VPP" scenario provides the capability to +The "Openstack - OpenDaylight - FD.io" scenario provides the capability to realize a set of use-cases relevant to the deployment of NFV nodes instantiated by means of an Openstack orchestration system on FD.io/VPP enabled compute nodes. The role of the Opendaylight network controller in this integration is @@ -74,22 +74,30 @@ servers: * 1 Jumphost hosting the APEX installer - running the Undercloud * 1 Controlhost, which runs the Overcloud as well as OpenDaylight as a network controller - * 2 or more Computehosts - -.. image:: FDS-odl_l3-overview.png - -Tenant and public networking leverages FD.io/VPP. VPP binds to both, the tenant -networking interface as well as to the public networking interface on the -compute and control nodes. The OpenDaylight network controller is used to setup -and manage layer 2 and layer 3 networking for the scenario - with Group Based -Policy (GBP) being the key component. Tenant networking can either leverage -VXLAN (in which case a full mesh of VXLAN tunnels is created) or VLANs. Layer 3 -connectivity is provided by using VPP as a "distributed virtual router". - -The picture below gives an example for distributed routing using VRFs between -tenant networks. - -.. image:: FDS-L3-DVR-example.png + * 2 or more Computehosts. One of the Computehosts also serves as + layer 3 gateway for the tenant networks. + +.. image:: FDS-odl_l3-noha-overview.png + +Tenant and public networking leverages FD.io/VPP. On one of the compute nodes, +VPP binds to both, the tenant networking interface as well as to the public +networking interface. This means that VPP is used for communication within +a tenant network, between tenant networks, as well as between a tenant network +and the Internet. Note that this setup slightly differs from the usual +centralized L3 setup with qrouter on the control node. This setup was chosen +to limit the configuration changes for the introduction of FD.io/VPP. The +OpenDaylight network controller is used to setup and manage layer 2 and +layer 3 networking for the scenario - with Group Based Policy (GBP) being the +key component. Tenant networking can either leverage VXLAN (in which case a +full mesh of VXLAN tunnels is created) or VLANs. + +The picture below shows an example setup with two compute and one control +node. Note that the external network is connected via compute node 0 through +VPP. VPP provides all layer 3 services which are provided in a "vanilla" +OpenStack deployment, including SNAT and DNAT, as well as north-south +and east-west traffic filtering for security purposes ("security groups"). + +.. image:: FDS-L3-noha-sample-setup.png Features of the scenario ------------------------ @@ -100,7 +108,7 @@ Main features of the "apex-os-odl_l3-fdio-noha" scenario: * Fast and scalable tenant networking using FD.io/VPP as forwarder * Layer 2 networking using VLANs or VXLAN, managed and controlled through OpenDaylight - * Layer 3 connectivitiy for tenant networks supplied in a distributed way + * Layer 3 connectivitiy for tenant networks supplied through FD.io/VPP. Layer 3 features, including security groups as well as floating IP addresses (i.e. NAT) are implemented by the FD.io/VPP forwarder * Manual and automatic (via DHCP) addressing on tenant networks @@ -122,86 +130,39 @@ and the Layer 3 scenario "apex-os-odl_l3-fdio-noha" share the same components. Here's a more detailed list of the individual software components involved: -**Openstack Neutron ML2 ODL Plugin**: Handles Neutron data base synchronization -and interaction with the southbound Openstack controller using HTTP. - -**OpenDaylight Neutron Nothbound & Neutron MD-SAL Entry Store**: Presents a -Neutron (v2) extended HTTP API servlet for interaction with Openstack Neutron. -It validates and stores the received Neutron data in the MD-SAL data store -against the Neutron yang model driven. - -**OpenDaylight Neutron Mapper**: The Neutron Mapper listens to Neutron data -change events and is responsible for using Neutron data in creating Group Based -Policy Data objects, e.g. GBP End-Points, Flood-Domains. A GBP End Point -represents a specific NFV/VM port and its identity as derived from a Neutron -Port. The mapped data is stored using the GBP End Point yang model and an -association between the GBP End-Point and its Neutron object is maintained in -the Neutron-GBP map. - -**OpenDaylight Group Based Policy (GBP) Entities store**: Stores for the GBP -data artifacts against the GBP YANG schemas. - -**Neutron Group Based Policy Map store**: Stores the bi-lateral relation -between an End-Point and its corresponding Neutron object. Neutron-GBP map; -keyed by Neutron object type, port, and Neutron UUID, gives the GBP End-Point, -Flood domain respectively. GBP-Neutron map keyed by GBP object type, end-point. - -**Neutron VPP Renderer Mapper**: The Neutron VPP Renderer Mapper listens to -Neutron Store data change events, as well as being able to access directly the -store, and is responsible for converting Neutron data specifically required to -render a VPP node configuration with a given End Point, e.g. the virtual host -interface name assigned to a vhostuser socket.. The mapped data is stored in -the VPP info data store. - -**VPP Info Store**: Stores VPP specific information regarding End-Points, Flood -domains with VLAN, etc. - -**GBP Renderer Manager**: The GBP Renderer Manager is the central point for -dispatching of data to specific device renderers. It uses the information -derived from the GBP end-point and its topology entries to dispatch the task of -configuration to a specific device renderer by writing a renderer policy -configuration into the registered renderer's policy store. The renderer manager -also monitors, by being a data change listener on the VPP Renderer Policy -States, for any errors in the application of a rendered configuration. - -**Renderer Policy Config Store**: The store's schema serves as the API between -the Renderer Manager and specific Renderers like the VPP Renderer. The store -uses a a YANG modeled schema to represent all end-point and associated GBP -policy data. - -**Topology Entries Store**: The yang model based MD-SAL topology store serves -two fundamental roles: 1. It maintains a topological representation of the GBP -End Points, in the context of customer networks. 2. It maintains an association -of each (VPP) compute node's physical interfaces to their neutron provider -network (e.g. The association between an ethernet interface and a Neutron -provider network). - -**VPP Renderer**: The VPP Renderer registers an instance for VPP nodes with the -Renderer Manager by means of inserting operational data into the Renderer -Policy config store. It acts as a listener on the Renderer Policy consumes via -the GBP Policy API data + the specific VPP End Point data, to drive the -configuration of VPP devices using NETCONF Services. -More specifically, the renderer generates: - - * vhost user port configuration that corresponds to the VM port configuration - * VPP bridge instances corresponding to the GBP flood domain - * port or traffic filtering configuration, in accordance with the GBP policy. - -The VPP Renderer also interacts with the Virtual Bridge Domain Service, by -means of the VBD store, in order to establish connectivity between VPP nodes in -a bridge domain. For this it uses the VPP device name, and the flood domain -data derived from the VPP Info and End-Point data respectively. For the -executed configuration operations it updates state in the Renderer policy state -store. - -**Virtual Bridge Domain (VBD) Store and Manager**: The virtual bridge domain -manager is responsible for configuring the VxLAN overlay tunnel infrastructure -to arrive at a desired bridged topology between multiple (VPP) compute nodes. -VDB configures VXLAN tunnels always into a full-mesh with split-horizon group forwarding applied on any domain facing tunnel interface (i.e. forwarding behavior will be that used for VPLS). - -**NETCONF Mount Point Service & Connector**: Collectively referred to as -Netconf Services, provide the NETCONF interface for accessing VPP configuration -and operational data stores that are represented as NETCONF mounts. +**Openstack Neutron ML2 OpenDaylight Plugin**: Handles Neutron data base +synchronization and interaction with the southbound controller using a REST +interface. + +**ODL GBP Neutron Mapper**: Maps neutron elements like networks, subnets, +security groups, etc. to GBP entities: Creates policy and configuration for +tenants (endpoints, resolved policies, forwarding rules). + +**ODL GBP Neutron VPP Mapper**: Maps Neutron ports to VPP endpoints in GBP. + +**ODL GBP Location Manager**: Provides real location for endpoints (i.e. Which +physical node an endpoint is connected to). + +**GBP Renderer Manager**: Creates configuration for Renderers (like e.g. +VPP-Renderer or OVS-Renderer). The GBP Renderer Manager is the central point +for dispatching of data to specific device renderers. It uses the information +derived from the GBP end-point and its topology entries to dispatch the task +of configuration to a specific device renderer by writing a renderer policy +configuration into the registered renderer's policy store. The renderer +manager also monitors, by being a data change listener on the VPP Renderer +Policy States, for any errors in the application of a rendered configuration. + +**GBP VPP Renderer Interface Manager**: Listens to VPP endpoints in the +Config DataStore and configures associated interfaces on VPP via HoneyComb. + +**GBP VPP Renderer Renderer Policy Manager**: Manages the creation of +bridge domains using VBD and assigns interfaces to bridge domains. + +**Virtual Bridge Domain Manager (VBD)**: Creates bridge domains (i.e. in case +of VXLAN creates full mesh of VXLAN tunnels, configures split horizon on +tunnel endpoints etc.). VDB configures VXLAN tunnels always into a full-mesh +with split-horizon group forwarding applied on any domain facing tunnel +interface (i.e. forwarding behavior will be that used for VPLS). **Virtual Packet Processor (VPP) and Honeycomb server**: The VPP is the accelerated data plane forwarding engine relying on vhost user interfaces @@ -209,20 +170,44 @@ towards Virtual Machines created by the Nova Agent. The Honeycomb NETCONF configuration server is responsible for driving the configuration of the VPP, and collecting the operational data. -**Rendered Policy State Store**: Stores data regarding the execution of -operations performed by a given renderer. - **Nova Agent**: The Nova Agent, a sub-component of the overall Openstack architecture, is responsible for interacting with the compute node's host Libvirt API to drive the life-cycle of Virtual Machines. It, along with the compute node software, are assumed to be capable of supporting vhost user interfaces. -The picture below show a basic end to end call flow for creating a Neutron -vhostuser port on VPP using a GBP renderer. It showcases how the different -component described above interact. - -.. image:: FDS-basic-callflow.jpg +The picture below shows the key components. + +.. image:: FDS-basic-components.jpg + +To provide a better understanding how the above mentioned components interact +with each other, the following diagram shows how the example of creating a +vhost-user port on VPP through Openstack Neutron: + +To create or update a port, Neutron will send a request to ODL Neutron +Northbound which contains the UUID, along with the host-id as "vpp" and +vif-type as "vhost-user". The GBP Neutron mapper turns the "Neutron speak" of +"ports" into the generic connectivity model that GroupBasedPolicy uses. +Neutron "ports" become generic "GBP Endpoints" which can be consumed by the +GBP Renderer Manager. The GBP Renderer Manager resolves the policy for the +endpoint, i.e. it determines which communication relationships apply to the +specific endpoint, and hands the resolution to a device specific renderer, +which is the VPP renderer in the given case here. VPP renderer turns the +generic policy into VPP specific configuration. Note that in case the policy +would need to be applied to a different device, e.g. an OpenVSwitch (OVS), +then an "OVS Renderer" would be used. VPP Renderer and the topology manager +("Virtual Bridge Domain" manager - i.e. VBD) cooperate to create the actual +network configuration. VPP Renderer configures the interfaces to the virtual +machines (VM), i.e. the vhost-user interface in the given case here and +attaches them to a bridge domain on VPP. VBD handles the setup of connectivity +between bridge domains on individual VPPs, i.e. it maintains the VXLAN tunnels +in the given case here. Both VPP Renderer as well as VBD communicate with the +device through Netconf/YANG. All compute and control nodes run an instance of +VPP and the VPP-configuration agent "Honeycomb". Honeycomb serves as a +Netconf/YANG server, receives the configuration commands from VBD and VPP +Renderer and drives VPP configuration using VPP's local Java APIs. + +.. image:: FDS-simple-callflow.png Scenario Configuration ====================== @@ -232,25 +217,51 @@ settings in the APEX configuration files. Those are typically found in /etc/opnfv-apex. File "deploy_settings.yaml": Choose Opendaylight as controller with version -"boron" and enable vpp as forwarder:: - - global_params: - ha_enabled: false +"carbon" and enable vpp as forwarder. "odl_routing_node" chooses the node +which is used as the layer 3 gateway for the tenant networks. +"odl_routing_node" is an optional parameter. If omitted, the VPP on the first +compute node will serve as layer 3 gateway:: deploy_options: sdn_controller: opendaylight sdn_l3: true - odl_version: boron - tacker: false - congress: false + odl_version: carbon + odl_routing_node: overcloud-novacompute-0 + tacker: true + congress: true sfc: false vpn: false vpp: true + dataplane: fdio + performance: + Controller: + kernel: + hugepages: 1024 + hugepagesz: 2M + intel_iommu: 'on' + iommu: pt + isolcpus: 1,2 + vpp: + main-core: 1 + corelist-workers: 2 + uio-driver: uio_pci_generic + Compute: + kernel: + hugepagesz: 2M + hugepages: 2048 + intel_iommu: 'on' + iommu: pt + isolcpus: 1,2 + vpp: + main-core: 1 + corelist-workers: 2 + uio-driver: uio_pci_generic Limitations, Issues and Workarounds =================================== -There are no known issues. +For specific information on limitations and issues, please refer to the APEX +installer release notes. References ========== @@ -260,4 +271,4 @@ References * Fast Data (FD.io): https://fd.io/ * FD.io Vector Packet Processor (VPP): https://wiki.fd.io/view/VPP * OpenDaylight Controller: https://www.opendaylight.org/ - * OPNFV Colorado release - more information: http://www.opnfv.org/colorado + * OPNFV Danube release - more information: http://www.opnfv.org/danube |
