Open-source NFV platform? OPNFV!

Transcription

1

2 BRKSDN-2333 Open-source NFV platform? OPNFV! Frank Brockners

3 Cisco Spark How Questions? Use Cisco Spark to communicate with the speaker after the session 1. Find this session in the Cisco Live Mobile App 2. Click Join the Discussion 3. Install Spark or go directly to the space 4. Enter messages/questions in the space cs.co/ciscolivebot#brkdcn Cisco and/or its affiliates. All rights reserved. Cisco Public

4 The 20 th century was about invention, the 21 st is about mashups and integration Toby Ford, FD.io Mini-Summit Sept, 2016 BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 5

5 NFV Stacks A User Perspective: Above and below The Line Service Model WorkFlow Topology App Intent Service Provisioning Workload Placement Service Configuration VM/Container Policy Service Monitoring Auto Healing Elastic Scaling Phys./virtual Network Control Group Policy, Chaining High-Performance Flexible Feature Paths Service/WF Life Cycle Manager Virtual Machine/Container Life Cycle Manager Network Controller; IO Abstraction & Feature Path VM Policy, Network Policy Service Provisioning, Service Configuration Service Chaining, Service Monitoring Auto Recovery, Elastic Scaling, Workload Placement, Service Assurance Hypervisor/Host//Container Compute Network Storage BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 6

6 Infrastructure Cloud Infra & Tooling PaaS OpenSource Building Blocks: ~ 2014/15 Application Layer / App Server PaaS platforms Network Data Analytics Orchestration VIM Management System Network Control Operating Systems IO Abstraction & Feature Path Hardware BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 7

7 Infrastructure Cloud Infra & Tooling PaaS OpenSource Building Blocks : Several New LF Projects Application Layer / App Server Additional PaaS platforms Network Data Analytics Orchestration VIM Management System Network Control * Operating Systems IO Abstraction & Feature Path Hardware BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 8

8 Composing the NO-STACK-WORLD Application Layer / App Server The No-Stack-Developer Network Data Analytics Orchestration VIM Management System Network Control Operating Systems IO Abstraction & Feature Path Compose Deploy Test Evolve Iterate OPNFV Hardware Evolve/Integrate/Install/Test OPNFV BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 9

9 CI/CD Infrastructure Cloud Infra & Tooling PaaS Assembling Platforms: OpenSource Building Blocks Application Layer / App Server Additional PaaS platforms Network Data Analytics Orchestration VIM Management System Network Control Operating Systems IO Abstraction & Feature Path Hardware Evolve/Integrate/Install/Test BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 10

10 Systems integration as an open community effort Cisco and/or its affiliates. All rights reserved. Cisco Public

11 Open Platform for NFV (OPNFV) facilitates the development and evolution of NFV components across various open source ecosystems. Through system level integration, deployment and testing, OPNFV creates a reference NFV platform to accelerate the transformation of enterprise and service provider networks. BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 12

12 OPNFV Danube Overview Virtual Network Functions Management and Orchestration Integration Testing New Features Compute Virtualization KVM Storage Virtualization OpenStack Ceph Network Virtualization OpenDaylight Alignment Installation Functional System NFV Features LXD ONOS Composition Performance OpenContrail Compute Storage Network Upstream Project Collaboration: FD.io Data Plane OVS Continuous Integration / Continuous Deployment DPDK Infrastructure Pharos Community Labs OPNFV Bare Metal Lab ODP Documentation Security BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 13

13 Create.Compose.Deploy.Test. Iterate Cisco and/or its affiliates. All rights reserved. Cisco Public

14 Create.Compose.Deploy.Test. Iterate Cisco and/or its affiliates. All rights reserved. Cisco Public

15 CI/CD Infrastructure Cloud Infra & Tooling PaaS Enhancing the Stack : Evolving/Developing while Integrating Application Layer / App Server Network Data Analytics NFV Telemetry Data Acquisition (VNF Event Stream, Barometer) Orchestration NFV deployment architectures (Multi-Site, EdgeNFV) MANO solutions Opera (Open-O), Orchestra (OpenBaton) VIM Management System Fault Management/Localization, Audit (Doctor, Pinpoint, Prediction, Bottlenecks, Inspector) Resource Control/Scheduling (Promise, Resource Scheduler (RS)) Model and Policy Driven Control (Copper, Movie, Models, Domino, Parser) Network Control Network Policy (FastDataStacks, Copper, PolicyTest) NFV services (service chaining, VPN,..) (VNFFG, SFC, SDNVPN) SDN Controller Performance Test (Cperf) Operating Systems, virt. Infra Hypervisor for NFV (KVM4NFV) IO Abstraction & Feature Path Dataplane Performance Test (Vsperf, Fastpath) Dataplane Evolution & Test (DPACC, OVSDPDK, FastDataStacks) Hardware Infra control & CI/CD (Pharos, Releng, Octopus) Integrate/Install Create Scenarios/Stacks (IPv6, FastDataStacks, Armband, Apex, Daisy, ONOSFW, OpenContrail,, Joid, Fuel, Compass) Installation, Upgrade (Apex, Joid, Fuel, Compass, Daisy, Escalator, Genesis) Documentation (OPNFVdocs) System Test & Security Audit System Testing (Yardstick, FuncTest, Qtip) Storage Performance Test (Storperf) Security (Moon) BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 16

16 Create.Compose: A typical workflow Requirement Architecture & Gaps Solution & review internally Reach & develop upstream Integrate, test and document Present demos, collect feedback BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 17

17 Example: OPNFV Doctor project Goal: Develop and build fault management and maintenance framework for high availability of Network Services running on top of virtualized infrastructure. Proposed with a very clear target / key feature: Immediate notification of unavailability of virtualized resources from VIM to Consumer Members: NEC (PTL: Ryota Mibu), AT&T, Cisco, Cloudbase Solutions, Corenova, Ericsson, Hephaex, Huawei, Intel, KDDI, KT, Nokia, NTT DOCOMO, Spirent, Sprint, Telecom Italia, Vmsec, ZTE BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 18

18 OPNFV Doctor project Timeline... ARNO (May/2015) - Requirement document BRAHMAPUTRA (March/2016) - Ceilometer Immediate Notification - Nova Mark Host Down - Functional test cases - PoC demo at OPNFV Summit - Documentation updates COLORADO (Sept/2016) DANUBE (March/2017) - Nova: - Neutron Port Status update Get valid server state - Inspector design guidelines - Integration of Congress - Performance profiler as Doctor Inspector - Documentation updates - Extended functional tests - PoC demo at OPNFV Summit and OpenStack Summit Barcelona - Documentation updates BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 19

19 Example: OPNFV FastDataStacks Project Create a new stack which significantly evolves networking for NFV: Introduce Solution Stacks with FD.io/VPP OpenStack ODL (Layer2) VPP OpenStack ODL (Layer3) VPP OpenStack VPP Work areas: OpenStack (ML2 driver: networking-vpp) ODL enhancements (GBP Neutron Mapper, VPP Renderer) FD.io - VPP/Honeycomb enhancements OPNFV Installer integration (APEX) OPNFV System-level testing Contributors: BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 20

20 OPNFV FastDataStacks project Timeline... Colorado 3.0 (Dec/2016) - Enhanced O/S- ODL(L2)-VPP stack (Infra complete: Neutron / GBP Mapper / GBP Renderer / VBD / Honeycomb / VPP) - Enhanced system-level testing - L2 networking using ODL (incl. east-west security groups), L3 networking uses qrouter/ovs - O/S-VPP (Infra: Neutron ML2-VPP / Networking-vpp-agent / VPP) - Automatic Install, Overlays: VLAN Danube 1.0 (March/2017) - Enhanced O/S- ODL(L3)-VPP stack (Infra complete: Neutron / GBP Mapper / GBP Renderer / VBD / Honeycomb / VPP) - L2 and L3 networking using ODL (incl. east-west security groups) BRKSDN-2333 Danube 2.0 (May/2017) - Enhanced O/S- ODL(L3/L2)-VPP stack: HA for OpenStack and ODL (clustering) Euphrates 1.0 (November/2017) - Enhanced O/S- ODL-VPP stack: DVR with VPP (pure L3 for eastwest and northsouth communication); Dynamic VXLANtunnel control via LISP 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 21

21 Building/evolving features means contributing upstream Team Up To Drive Required Change Upstream Swimming upstream is hard Swimming upstream is dangerous Team-up to increase the likelihood to succeed BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 22

22 Create.Compose.Deploy.Test. Iterate Cisco and/or its affiliates. All rights reserved. Cisco Public

23 Choice of existing and new components, System state definition Integrate & build* as a system Test on reference system(s) Install on reference system(s) Lab1 Lab2 Lab3 *OPNFV currently composes builts from existing artifacts (e.g. RPMs) rather than builds from source Lab1 Lab2 Lab3 OPNFV CI/CD Projects: Octopus, Releng BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 24

24 OPNFV CI/CD Project CI Google Cloud Storage Patch Verified +1/-1 Poll changes Clone repo Upload ISO Build new changes? yes Build Build Build Server Patchset Verification no Push Image Docker Hub BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 25

25 OPNFV CI/CD Platform CI Google Cloud Storage Docker Hub Docker Hub Labs connected 24x7 using JNLP Download ISO Pull Down Functest Image Pull Down Yardstick Image Initiate Run for certain Scenario Jumphost Deploy Functest Yardstick Pharos POD (SUT) BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 26

26 OPNFV Development Workflow Project Team OPNFV Gerrit OPNFV CI/Test OPNFV Release Upstream Gerrit Upstream CI/Test Upstream Release Requirement Patch Fast feedback Test Test Release Document Downstream Test Release Fast Long development cycle Fast Downstream feedback, will < delay 1 dayto next release OPNFV Slow feedback, specific > issues 5 months can be tested/detected OPNFV specific in issues time; cannot be Correction tested/detected in same in time release 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public

27 Cross Community CI (XCI) contribute contribute contribute contribute pull from master, deploy, test and/or verify patchset, post feedback BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 28

28 Deploy Develop - Test Pharos Project: Community Labs Goals: Facilitate collaborative testing Provide developers with substantial resources Ensure OPNFV applicability across architectures, environments and vendors Create more robust, interoperable releases BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 29

29 Infrastructure Distributed Pharos Labs Facilitate collaborative testing Ensure OPNFV applicability across architectures, environments and vendors Create more robust, interoperable releases BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 30

30 Pharos Architecture BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 31

31 Infrastructure Homogeneity Lab as a Service Resources booked via dashboard BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 32

32 CI/CD Dynamic Resource Allocation Dedicated Resources Installer X Scenario A Dedicated Resources Resources Dynamically Allocated Installer Y Scenario B BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 33

33 OPNFV testing and integration project NFV orchestration Virtualized Network Functions VNF VNF VNF VNF Cloud Management Infrastructure Virtualization Compute Storage Network BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 34

34 Types of testing Functional Performance Stress Security Infrastructure Features Components VNF MANO Network Storage Compute Virtual layer Traffic gen. Network Storage Compute Upgradability Backup&Restore... BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 35

35 OPNFV Test Ecosystem BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 36

36 Test: Functional Testing: Example Project FuncTest vping SSH Verify private & public connectivity vping userdata Verify nova-metadata service and private connectivity Tempest test OpenStack native tests (200+ tests) Rally bench tests Benchmark the OpenStack deployment ODL suite Robot framework, ODL functional testing ONOS suite TestON framework vims OpenSource solution by Clearwater Promise Resource reservation and management project Doctor Fault management and maintenance project BGPVPN OpenStack Neutron BGPVPN project integration BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 37

37 System-Level testing: Example Project YardStick ETSI GS NFV-TST001 Chapter 6 Pre-Deployment validation of NFV infrastructure Pre-Deployment validation of NFVI 1. Define infrastructure 2. Identify VNF type 3. Select Test Cases Control plane/user plane Main components Real time constraints Hardware Metrics YardStick Test-Framework Configure Deploy - System under Test (SUT) - Deployment rules - Generate template or - External templates 4. Execute Tests 5. Collect Results Test cases 1 N Test stimuli Test environment Test - Execute benchmarks - Trigger External scripts for SUT disturbances - Collect results 6. Integrate VNF application Validate - Graphical results - Assertion BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 38

38 Compute Network Performance/Speed Capacity/Scale Reliability/Availability - Latency for random memory access - Latency for cache read/write operations - Processing speed (instructions per second) -Throughput for random memory access (bytes per second) - Throughput per NFVI node (frames/byte per second) - Throughput provided to a VM (frames/byte per second) - Latency per traffic flow - Latency between VMs - Latency between NFVI nodes - Packet delay variation (jitter) between VMs - Packet delay variation (jitter) between NFVI nodes - Number of cores and threads - Available memory size - Cache size - Processor utilization (max, average, standard deviation) - Memory utilization (max, average, standard deviation) - Cache utilization (max, average, standard deviation) - Number of connections - Number of frames sent/received - Maximum throughput between VMs (frames/byte per second) - Maximum throughput between NFVI nodes (frames/byte per second) - Network utilization (max, average, standard deviation) - Number of traffic flows - Processor availability (Error free processing time) - Memory availability (Error free memory time) - Processor mean-time-to-failure - Memory mean-time-to-failure - Number of processing faults per second - NIC availability (Error free connection time) - Link availability (Error free transmission time) - NIC mean-time-to-failure - Network timeout duration due to link failure - Frame loss rate Storage - Sequential read/write IOPS - Random read/write IOPS - Latency for storage read/write operations - Throughput for storage read/write operations - Storage/Disk size - Capacity allocation (block-based, object-based) - Block size - Maximum sequential read/write IOPS - Maximum random read/write IOPS - Disk utilization (max, average, standard deviation) - Disk availability (Error free disk access time) - Disk mean-time-to-failure - Number of failed storage read/write operations per second BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 39

39 Test Dashboards BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 40

40 Test Dashboards BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 41

41 Operations and Troubleshooting support: Operating OpenStack Networking is not that simple oting-neutron-physical-and-virtual-networks ( Operations War Stories from OS Summit) BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 42

42 OPNFV Project Calipso Solves virtual networking discovery issues Monitoring support for day2 net operations Discover, Analyze, Monitor, Visualize Substantially lowering administration overhead for maintenance and troubleshooting cycles in cloud environments ( < 90% for discovery, analysis and monitoring of v-nets) Highly Customized and Modeled calipso.io and BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 43

43 Calipso BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 44

44 Calipso BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 44

45 Calipso feature set today ( out of the box ): Auto-discovery of virtual networking components, their detailed data for K8S and OpenStack Auto-discovery and end-to-end monitoring for virtual-physical-virtual for ACI/APIC use-cases Analysis of inter-connections and dependency (links) with embedded impact analysis Monitoring of virtual networking objects and links for health and status, some statistics Visualization of topologies, statuses, settings, browsing tree and an embedded search engine. API for Eco-system integrations. Support multi-region and multi-tenancy in multiple OpenStack, K8s environments Integrated monitoring and a plug-in framework, ready for many type of devices A unified all-in-one Installer for simple deployment. Modeled, Federated and customized, model driven design and micro-services architectures. BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 46

46 OPNFV NFVbench: Evaluating NFVI performance Full-Stack NFV blackbox NFV-benchmark: NFVbench tool Different connection scenarios (PVP, PVVP,..) Different traffic patterns NDR/PDR detection Different VNFs UCS-B Fabric Switch NIC NIC NIC VPP VPP VNF1 Compute 1 HoneyComb Nova VNF2 HoneyComb NFVbench (with Trex) Nova Compute 2 Compare a FastDataStack to legacy stacks with other virtual switches, e.g. OVS NIC VPP HoneyComb Controller UnderCloud Controller NIC OverCloud Jumphost FastDataStack full NFV stack deployment BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 47

47 An NFV Solution Stack is only as good as its foundation BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 48

48 OPNFV Releases 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public

49 OPNFV Release History June 4, 2015 March 1, 2016 September 26, 2016 April 4, 2017 Baseline foundation of components necessary to build an NFV platform from upstream components Massively parallel simultaneous release process. Advancements in infrastructure, processes, and upstream collaboration Platform support for NFV applications and key improvements in services and support Brings together end-to-end networking stacks, including MANO, data plane acceleration, and architecture advancements. BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 50

50 OPNFV Danube (April 2017) The only platform that brings together elements across multiple end-to-end open networking stacks Foundational support and introduction of capabilities for MANO including integration with ONAP (Open-O) Enhanced DevOps automation and testing methodologies, including performance and benchmarking test suites Architectural improvements including greater network control flexibility, HA, and multisite improvements Focus on NFV performance including acceleration of the data plane via FD.io integration and enhancements to OVS-DPDK and KVM Feature enrichment and maturity in core NFVI/VIM functionality such as IPv6, Service Function Chaining (SFC), L2 and L3 Virtual Private Network (VPN), fault management and analysis, and a continued commitment to support multiple hardware architectures BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 51

51 OPNFV Euphrates (October 2017) Introduces Kubernetes container orchestration engine as an alternative VIM, and with SDN controllers including basic functional and performance testing capabilities. A containerized version of OpenStack simplifies OpenStack lifecycle management. Integrates cross-community continuous integration (XCI), where the OPNFV CI pipeline integrates the latest upstream code, reducing feedback time on a new feature or bug-fix from months to days. Delivers an extensive set of tools to test the NFVI/VIM layer (i.e. NFV cloud), VNFs and complete network services. With two new projects: SampleVNF and NFVbench Includes new carrier-grade features through a new project, Calipso, where operators gain visibility to their complex virtual networks and access powerful service assurance framework. Performance improvements on the ARM architecture and in L3 performance with FD.io; new security, SFC, FD.io and EVPN features. Focus on Analytics and Benchmarking Barometer (CollectD for OPNFV) VNF Event Stream (VES) Bamboo (PNDA.io for OPNFV) Virtual Infrastructure Networking Assurance (Calipso) NFVbench (Full Stack NFVI one-shot benchmarking) Yardstick Qtip BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 52

52 OPNFV and ONAP: Thoughts on alignment Develop.Integrate: Management & Orchestration: XCI: Deploy.Test (OPNFV, ODL, ONAP, FD.io, PNDA.io,...) Develop.Integrate: NFV-I BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 53

53 Example: The OPNFV FastDataStacks project 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public

54 Foundational Assets For NFV Infrastructure: A stack is only as good as its foundation Virtual Forwarder Feature rich, high performance, highly scalable virtual switch-router Leverages hardware accelerators Runs in user space Modular and easy extensible Forwarder Diversity: Hardware and Software Virtual Domains link and interact with physical domains Domains and Policy Connectivity should reflect business logic instead of physical L2/L3 constructs Service Model WorkFlow Topology App Intent Service/WF Life Cycle Manager Virtual Machine/Container Life Cycle Manager Network Controller Forwarder Switch/Router BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 55

55 Networking Foundation for NFV Infrastructure Choices VPP Highly scalable, high performance, extensible virtual forwarder OpenDaylight Extensible controller platform Decouple business logic from network constructs: Group Based Policy as mediator between business logic and network constructs Support for a diverse set of network devices Clustering for HA BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 56

56 Introducing Vector Packet Processor - VPP VPP is a rapid packet processing development platform for highly performing network applications 14+ MPPS, single core Multimillion entry FIBs 480Gbps bi-dir on 24 cores NC/Y REST... Management Agent Runs on commodity CPUs and leverages DPDK Creates a vector of packet indices and processes them using a directed graph of nodes resulting in a highly performant solution. Runs as a Linux user-space application Ships as part of both embedded & server products, in volume; Active development since 2002 Packet Processing: VPP Network IO See also: FD.IO (The Fast Data Project) BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 57

57 VPP Performance at Scale Phy-VS-Phy [Gbps]] [Mpps] Gbps zero frame loss 200Mpps zero frame loss IPv6, 24 of 72 cores IPv4+ 2k Whitelist, 36 of 72 cores Zero-packet-loss Throughput for 12 port 40GE 1518B 64B 1518B 64B [Gbps]] [Mpps] B IMIX => 342 Gbps,1518B => 462 Gbps B => 238 Mpps 64B Hardware: Cisco UCS C460 M4 Intel C610 series chipset 4 x Intel Xeon Processor E v3 (18 cores, 2.5GHz, 45MB Cache) 2133 MHz, 512 GB Total 9 x 2p40GE Intel XL x 40GE = 720GE!! Latency 18 x 7.7trillion packets soak test Average latency: <23 usec Min Latency: 7 10 usec Max Latency: 3.5 ms Headroom Average vector size ~24-27 Max vector size 255 Headroom for much more throughput/features NIC/PCI bus is the limit not vpp BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 58

58 NFV is about forwarding delay: VPP means low delay Low long-term max packet delay with FD.io VPP 0.007ms < 0.023ms < 3.5ms Average Delay < 23 usec Min Delay < 10 usec Max Delay < 3.5 msec Other vswitches >120ms long term max delay Tests environment stock Ubuntu LTS Kernel: generic (no Linux tuning) Cisco UCS C460 M4 (4 x Intel Xeon Processor E v3 (18 cores, 2.5GHz, 45MB Cache), 9 x 2p40GE Intel XL710) BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 59

59 VPP Features as of Release Hardware Platforms Routing Switching Network Services Pure Userspace - X86,ARM 32/64, Power, Raspberry Pi Interfaces DPDK/Netmap/AF_Packet/TunTap Vhost-user - multi-queue, reconnect, Jumbo Frame Support Language Bindings C/Java/Python/Lua Tunnels/Encaps GRE/VXLAN/VXLAN-GPE/LISP-GPE/NSH IPSEC Including HW offload when available MPLS MPLS over Ethernet/GRE Deep label stacks supported IPv4/IPv6 14+ MPPS, single core Hierarchical FIBs Multimillion FIB entries Source RPF Thousands of VRFs Controlled cross-vrf lookups Multipath ECMP and Unequal Cost Segment Routing SR MPLS/IPv6 Including Multicast LISP LISP xtr/rtr L2 Overlays over LISP and GRE encaps Multitenancy Multihome Map/Resolver Failover Source/Dest control plane support Map-Register/Map-Notify/ RLOC-probing VLAN Support Single/ Double tag L2 forwd w/efp/bridgedomain concepts VTR push/pop/translate (1:1,1:2, 2:1,2:2) Mac Learning default limit of 50k addr Bridging Split-horizon group support/efp Filtering Proxy Arp Arp termination IRB - BVI Support with RouterMac assigmt Flooding Input ACLs Interface cross-connect L2 GRE over IPSec tunnels Security Mandatory Input Checks: TTL expiration header checksum L2 length < IP length ARP resolution/snooping ARP proxy SNAT Ingress Port Range Filtering Per interface whitelists Policy/Security Groups/GBP (Classifier) DHCPv4 client/proxy DHCPv6 Proxy MAP/LW46 IPv4aas MagLev-like Load Identifier Locator Addressing NSH SFC SFF s & NSH Proxy LLDP BFD Policer Multiple million Classifiers Arbitrary N-tuple Inband ioam Telemetry export infra (raw IPFIX) ioam for VXLAN-GPE (NGENA) SRv6 and ioam co-existence ioam proxy mode / caching ioam probe and responder Monitoring Simple Port Analyzer (SPAN) IP Flow Export (IPFIX) Counters for everything Lawful Intercept BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 60

60 Rapid Release Cadence ~3 months Release: VPP, Honeycomb, NSH_SFC, ONE Release: VPP, Honeycomb, NSH_SFC, ONE Release: VPP, Honeycomb, NSH_SFC, ONE Release: VPP, Honeycomb, NSH_SFC, ONE NewFeatures New Features New Features New Features Enhanced LISP support for L2 overlays Multitenancy Multihoming Re-encapsulating Tunnel Routers (RTR) support Map-Resolver failover algorithm New plugins for SNAT MagLev-like Load Identifier Locator Addressing NSH SFC SFF s & NSH Proxy Port range ingress filtering Dynamically ordered subgraphs Hierarchical FIB Performance Improvements DPDK input and output nodes L2 Path IPv4 lookup node IPSEC Performance SW and HW Crypto Support HQoS support Simple Port Analyzer (SPAN) BFD, ACL, IPFIX, SNAT L2 GRE over IPSec tunnels LLDP LISP Enhancements Source/Dest control plane L2 over LISP and GRE Map-Register/Map-Notify RLOC-probing Flow Per Packet VPP Userspace Host Stack TCP stack DHCPv4 & DHCPv6 relay/proxy ND Proxy SNAT CGN: port allocation & address pool CPE: External interface NAT64, LW46 Segment Routing SRv6 Network Programming SR Traffic Engineering SR LocalSIDs Framework to expand LocalSIDs w/ plugins ioam UDP Pinger IOAM as type 2 metadata in NSH Anycast active server selection IPFIX Improvements (IPv6) VPP Userspace Host Stack Improved Linux Host Stack Compatibility Improved loss recovery (RFC5681, RFC6582, RF6675) Basic implementation of Eifel detection algorithm (RFC3522) Interfaces memif: IP mode, jumbo frames, multi queue Network Features MPLS Multicast FIB BFD FIB integration NAT64 support GRE over IPv6 LISP - NSH support - L2 ARP ACL multi-core suuport GTP-U support BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 61

61 Control & configuration plane for VPP: Honeycomb A generic, model driven (YANG) management agent Northbound interfaces: RESTCONF NETCONF Generic data processing layer (transaction management, validation) Data processing layer: Pipeline processing data from northbound interfaces down to translation layer Translation layer: Invoked by above layer to handle configuration updates or when polling operational state from VPP. Specific translation code lives in this layer in a form of extensions/plugins Extensible, pluggable translation layer to handle resulting data in any way HoneyComb BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 62

62 HoneyComb Key Features Key Features: Config, Operational, Context data processing, Persistence & Reconciliation, Rollback, CRUD operations on data, Notifications, Simple JSON plugin configuration Infrastructure: Utilizes existing Opendaylight components and tools: Yangtools, MD-SAL, Netconf, Restconf Does not use ODL s config subsystem, OSGi or any other container Wired into a special data processing pipeline Running in a JVM Reduced Footprint (compared to ODL) < 100Mb RAM, < 10sec Startup time, < 30Mb of disk BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 63

63 Group Based Policy in Open Daylight Group policy for generic end points Application-focused policy expressions: Policies mirror application semantics. Capture policy requirements without detailed knowledge of networking. Improved automation: Grouping constructs allow higher level automation tools to easily manipulate groups of network endpoints simultaneously. Consistent policy by grouping end points and applying policy to groups Extensible because of implementation independence, hence applicable to policy for connectivity, security, L4-7, QoS, etc. See also: BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 64

64 OK. Ready to integrate! BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 65

65 Evolving The OPNFV Scenario Set OPNFV uses scenarios (i.e. compositions of features and their configuration) as key release vehicle Let s create a new stack which significantly evolves networking for NFV Introduce Scenarios with VPP for OPNFV os-nosdn-fdio-noha, os-nosdn-fdio-ha os-odl_l2-fdio-noha, os-odl_l2-fdio-ha os-odl_l3-fdio-noha, os-odl_l3-fdio-ha Category Install Tools VM Control Network Control Hypervisor Components in OPNFV Apex, Compass, Fuel, Juju OpenStack OpenDaylight, ONOS, OpenContrail KVM, KVM4NFV Forwarder OVS, OVS-DPDK + VPP BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 66

66 OPNFV FastDataStacks (FDS) Integrate VPP into existing OPNFV scenarios Initial scenarios OpenStack ODL (Layer2) VPP OpenStack ODL (Layer3) VPP OpenStack VPP... Install Tools VM Control Network Control Apex OpenStack OpenDaylight L2 Apex OpenStack OpenDaylight L3 Apex OpenStack Diverse set of contributors: Hypervisor KVM KVM KVM Forwarder VPP VPP VPP BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 67

67 FastDataStacks Component Development: Direct Integration of VPP with OpenStack OpenStack Networking-vpp ML2 driver FD.io VPP Enhancements CSIT VPP component tests OPNFV Overall System Composition Integration into CI/CD Installer: Integration of VPP into APEX System Test: FuncTest and Yardstick system test application to FDS... Neutron ML2-driver: networking-vpp etcd ML2-agent VPP DPDK System Install (APEX) System Test (FuncTest, Yardstick) See also: FDS Architecture: BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 68

68 VPP Agent VPP Agent Networking-vpp: overall architecture Neutron Server ML2 VPP Mechanism Driver journaling VM VM VM HTTP/json VM VM VM vhostuser VPP vhostuser VPP dpdk Compute Node dpdk Compute Node vlan / flat network BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 69

69 Tenant network i/f Tenant network i/f FastDataStacks: OS FD.io Example: 3 node setup: 1 x Controller, 2 x Compute Internet Controlnode-0 External network i/f OVS (br-ex) Computenode-0 OpenStack Services Neutron Server ML2 VPP Mechanism Driver qrouter (NAT) tap Computenode-1 ML2 Agent DHCP tap Bridge Domain VPP ML2 Agent ML2 Agent Tenant network i/f VM 1 Bridge Domain Bridge Domain vhostuser vhostuser VM 2 VPP VPP VLAN / Flat network BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 70

70 Networking-vpp: current feature set Network types VLAN: supported since version VXLAN-GPE: supported since version Port types VM connectivity done using fast vhostuser interfaces TAP interfaces for services such as DHCP Security Security-groups based on VPP stateful ACLs Port Security can be disabled for true fastpath Role Based Access Control and secure TLS connections for etcd JSON Web Token (RFC 7519) with X.509 Certificate Tap as a Service (scheduled for 17.10) Layer 3 Networking North-South Floating IP North-South SNAT East-West Internal Gateway Robustness If Neutron commits to it, it will happen Component state resync in case of failure: recovers from restart of Neutron, the agent and VPP BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 71

71 VPP Agent Networking-vpp: port creation process networking-vpp/nodes/vpp- rocks/ports/c367e21f-ae b87d- 2e c6 1 5 Request Notification VM {"allowed_address_pairs": [], "segmentation_id": 194, "mtu": 1500, "binding_type": "plugtap", "physnet": "physnet", "mac_address": "fa:16:3e:03:ce:ff", "port_security_enabled": false, "fixed_ips": [{"subnet_id": "006fce a695-c3caa140fff7", "ip_address": " "}, {"subnet_id": "81b2fbdc-c350-4f35-9b9b- 909cf33a4426", "ip_address": "fd59:3bf6:c35d:0:f816:3eff:fe03:ceff"}], "network_type": "vlan", "security_groups": []} Neutron Server ML2 VPP Mechanism Driver vhostuser VPP dpdk Compute Node 4 /networking-vpp/state/vpp-rocks/ports/d2069a46-3a47-4ec7-94fb- 3b1bcd4c6dc0 {"net_data": {"segmentation_id": null, "if_physnet": "tap-2", "bridge_domain_id": 3, "if_uplink_idx": [3], "network_type": "flat", "physnet": "physnet"}, "bind_type": "plugtap", "ext_tap_name": "tapd2069a46-3a", "mac": "fa:16:3e:5d:fe:c4", "bridge_name": "br-d2069a46-3a", "int_tap_name": "vppd2069a46-3a", "iface_idx": 6} 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public

72 FastDataStacks Component Development Controller-based integration of VPP with OpenStack OpenDaylight GBP Neutron Mapper GBP Renderer Manager enhancements VPP Renderer Virtual Bridge Domain Mgr / Topology Manager FD.io HoneyComb Enhancements VPP Enhancements CSIT VPP component tests OPNFV Overall System Composition Integration into CI/CD Installer: Integration of VPP into APEX System Test: FuncTest and Yardstick system test application to FDS VPP renderer Neutron Neutron NorthBound GBP Neutron Mapper Topology Mgr - VBD Honeycomb (Dataplane Agent) System Install (APEX)... VPP REST GBP Renderer Manager Netconf/YANG DPDK System Test (FuncTest, Yardstick) See also: FDS Architecture: BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 73

73 Tenant network i/f Tenant network i/f FastDataStacks: OS ODL(L2) FD.io Example: 3 node setup: 1 x Controller, 2 x Compute Internet Controlnode-0 External network i/f OVS (br-ex) Computenode-0 OpenStack Services qrouter (NAT) Network Control Computenode-1 tap HoneyComb DHCP tap Bridge Domain VPP HoneyComb HoneyComb Tenant network i/f VM 1 Bridge Domain VXLAN VXLAN VXLAN Bridge Domain vhostuser vhostuser VM 2 VPP VPP BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 74

74 Tenant network i/f Tenant network i/f FastDataStacks: OS ODL(L3) FD.io Example: 3 node setup: 1 x Controller, 2 x Compute Internet Controlnode-0 External network i/f Computenode-0 OpenStack Services Network Control Computenode-1 HoneyComb DHCP tap Bridge Domain VPP HoneyComb HoneyComb Tenant network i/f VM 1 Bridge Domain VXLAN VXLAN VXLAN Bridge Domain vhostuser vhostuser VM 2 VPP VPP BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 75

75 Example: Creating a Neutron vhostuser port on VPP Neutron Neutron NorthBound POST PORT (id=<uuid>, host_id=<vpp>, vif_type=vhostuser) Update Port Map Port to GBP Endpoint GBP Neutron Mapper Update/Create Policy involving GBP Endpoint GBP Renderer Manager Resolve Policy Apply policy, update nodes VPP Renderer VM Netconf/ YANG vhostuser configure interfaces over Netconf Honeycomb VPP 1 VXLAN Tunnel Topology Manager (vbd) Netconf/ YANG Honeycomb VPP 2 Bridge domain and tunnel config Configure bridge domain on nodes over NetConf BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 76

76 Let s install an OPNFV Solution Stack: Using APEX/TripleO Installer to install a FastDataStack BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 78

77 Apex Installer Apex is an OPNFV Installation and Deployment tool based on the RDO Project's Triple-O OpenStack installation tool Triple-O is an image based life cycle deployment tool that is a member of the OpenStack Big Tent Governance BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 79

78 APEX Install & Build Jump Host Setup Apex ISO CentOS Virtualization Host + Apex RPMs Git Repo (local build) Build artifacts: CentOS ISO Apex RPMs Undercloud libvirt image Glance overcloud images Undercloud VM OpenStack APEX / TripleO Overcloud Controller OpenStack Services OpenDaylight FD.io: Honeycomb, VPP qrouter OVS (br-ext) Overcloud Compute 1..N Nova compute NIC 0 NIC 2 NIC 0 NIC 1 NIC 2 NIC 0 Admin Public Tenant FD.io: Honeycomb, VPP NIC 1 BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 80

79 Apex: Deployment Jump Host Jump Host Libvirt image Undercloud VM BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 82

80 Apex: Deployment Apex: Deployment Jump Host Undercloud (VM) Glance images OPNFV (overcloud) BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 83

81 Apex: Deployment Jump Host Undercloud (VM) OPNFV OPNFV Control OPNFV Control Control OPNFV Compute OPNFV Compute Overcloud BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 84

82 Apex: Deployment BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 85

83 APEX deploy command: 3 config files nohup opnfv-deploy --debug -n /etc/opnfvapex/network_settings.yaml -d /etc/opnfv-apex/os-odl_l2- fdio-noha.yaml -i /etc/opnfv-apex/inventory.yaml & BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 86

84 Example: /etc/opnfv-apex/os-odl_l2-fdio-noha.yaml global_params: ha_enabled: false deploy_options: sdn_controller: opendaylight #Use OpenDaylight as network controller sdn_l3: false odl_version: boron #OpenDaylight Boron is required for FDS tacker: true congress: true sfc: false vpn: false vpp: true #use VPP dataplane: fdio #use VPP for communication between compute nodes performance: Controller: kernel: hugepages: 1024 #VPP setups default to 1024 hugepagesz: 2M intel_iommu: 'on' iommu: pt Compute: kernel: hugepagesz: 2M hugepages: 2048 #Should be as high as possible for the compute nodes intel_iommu: 'on' #enable IOMMU iommu: pt #IOMMU pass through mode disable IOMMU for the hypervisor BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 87

85 Deployment done! tail -5 nohup.out == Keystone users == Warning keystonerc not sourced Undercloud IP: , please connect by doing 'opnfv-util undercloud' Overcloud dashboard available at INFO: Post Install Configuration Complete BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 88

86 Ready... BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 90

87 Let s use our FastDataStack OPNFV Solution BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 91

88 Running a FastDataStack (1/2) 1. Get an image wget -O /tmp/cirros x86_64-disk.img 2. Upload the image into Glance and make it support hugepages glance image-create --name "cirros x86_64" --file /tmp/cirros x86_64-disk.img --disk-format qcow2 --container-format bare --visibility public--progress glance image-update --property hw_mem_page_size=large ad74564d-fd22-414b-9fa3-619e87f781a9 3. Create a flavor with hugepages enabled nova flavor-create --is-public true opnfv nova flavor-key opnfv set hw:mem_page_size=large 4. Create a Neutron network neutron net-create vxlannet --provider:network_type vxlan BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 92

89 Running a FastDataStack (2/2) 5. Assign a Subnet to your network neutron subnet-create vxlannet /24 --name vxlansubnet 6. Create a Neutron Router neutron router-create r1 neutron router-interface-add r1 subnet=vxlansubnet neutron router-gateway-set r1 external 7. (Optional) Create Ports for your VMs neutron port-create vxlannet --name vm-compute-0-1-port neutron port-create vxlannet --name vm-compute-1-1-port (booting the VMs will do this for you) 8. Boot your VMs nova boot --image cirros x86_64 --flavor opnfv --nic port-id=`neutron port-list grep vm-compute-0-1-port cut -f 2 -d " "` vm-compute availability-zone nova:overcloud-novacompute-0.localdomain.com nova boot --image cirros x86_64 --flavor opnfv --nic port-id=`neutron port-list grep vm-compute-1-1-port cut -f 2 -d " "` vm-compute availability-zone nova:overcloud-novacompute-1.localdomain.com BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 93

90 Network Setup Complete: Example OpenStack Perspective: Check your ports ~]# neutron port-list id name mac_address fixed_ips fbf0-5a9b- fa:16:3e:8c:8f:f {"subnet_id": "1e9a 48f2-b88f- 2 5a26-478e-4837-a659 e115093d2594 -ca7f6105a7e3", "ip_address": " "} 9bf477cb-a016 vm- fa:16:3e:db:e5:0 {"subnet_id": "1e9a -407e-a97a- compute-0-1-port 2 5a26-478e-4837-a659 7f15b35c3bad -ca7f6105a7e3", "ip_address": " "} b504c2a4-6c6c-41 vm- fa:16:3e:99:24:4 {"subnet_id": "1e9a c0-b011-a3e747c6 compute-1-1-port 1 5a26-478e-4837-a ca7f6105a7e3", "ip_address": " "} f70a5d22-e6cd- fa:16:3e:3f:50:e {"subnet_id": "1e9a 4c36-8b7c- 4 5a26-478e-4837-a659 66ddd14c469f -ca7f6105a7e3", "ip_address": " "} DHCP tap port Port for VM #1 Port for VM #2 qrouter tap port BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 94

91 Network Setup Complete: Example ODL Active ODL Perspective: Network Topology Active Network Topology ~]# curl -XGET -u admin:admin python -mjson.tool { "network-topology": { "topology": [ { "link": [ { "destination": { "dest-node": "overcloud-novacompute-0.opnfvapex.com", "dest-tp": "vxlan_tunnel5" }, "link-id": "overcloud-novacompute-1.opnfvapex.com-1-overcloud-novacompute-0.opnfvapex.com", "source": { "source-node": "overcloud-novacompute-1.opnfvapex.com", "source-tp": "vxlan_tunnel4" }, "vbridge-topology:tunnel": "vxlan_tunnel4" }, { "destination": { "dest-node": "overcloud-controller-0.opnfvapex.com", "dest-tp": "vxlan_tunnel4" }, "link-id": "overcloud-novacompute-0.opnfvapex.com-1-overcloud-controller-0.opnfvapex.com", "source": { "source-node": "overcloud-novacompute-0.opnfvapex.com", "source-tp": "vxlan_tunnel4" }, "vbridge-topology:tunnel": "vxlan_tunnel4" }, [... abbreviated...] BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 95

92 Network Setup Complete: Example HoneyComb Perspective: Interfaces ~]# curl -XGET -v -u admin:admin python -m json.tool % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed :--:-- --:--:-- --:--:-- 0* About to connect() to localhost port 8182 (#0) * Trying ::1... * Connection refused * Trying * Connected to localhost ( ) port 8182 (#0) * Server auth using Basic with user 'admin' > GET /restconf/config/ietf-interfaces:interfaces/ HTTP/1.1 > Authorization: Basic YWRtaW46YWRtaW4= > User-Agent: curl/ > Host: localhost:8182 > Accept: */* > < HTTP/ OK < Date: Sun, 18 Sep :29:10 GMT < Content-Type: application/yang.data+json < Transfer-Encoding: chunked < Server: Jetty( v ) < { [data not shown] C0 C k 0 --:--:-- --:--:-- --:--:-- 162k * Connection #0 to host localhost left intact { "interfaces": { "interface": [ { "description": "neutron port", "enabled": true, Ctl "link-up-down-trap-enable": "enabled", "name": "neutron_port_f70a5d22-e6cd-4c36-8b7c-66ddd14c469f", "type": "v3po:tap", "v3po:l2": { "bridge-domain": "fea5f ff5-a4a1-e6404e6b6407" }, "v3po:tap": { "mac": "fa:16:3e:3f:50:e4", "tap-name": "qr-f70a5d22-e6" } QR DHCP }, BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 96

93 Network Setup Complete: Example VPP perspective - vppctl show int addr [root@overcloud-controller-0 ~]# vppctl show int addr TenGigabitEthernet7/0/0 (up): /24 local0 (dn): tap-0 (up): l2 bridge bd_id 1 shg 0 tap-1 (up): l2 bridge bd_id 1 shg 0 vxlan_tunnel0 (up): l2 bridge bd_id 1 shg 1 vxlan_tunnel1 (up): l2 bridge bd_id 1 shg 1 BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 97

94 VMs up and running Ready to Ping ~]# nova list ID Name Status Task State Power State Networks a2a66d-b73c-470d bd0a9076e61 vm-compute-0-1 ACTIVE - Running vxlannet= cf0-ac93-9f2ba3dcf984 vm-compute-1-1 ACTIVE - Running vxlannet= BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 98

95 OPNFV FastDataStacks project Timeline... Colorado 1.0 (Sep/2016) - Base O/S- ODL(L2)-VPP stack (Infra: Neutron / GBP Mapper / GBP Renderer / VBD / Honeycomb / VPP) - Automatic Install - Basic systemlevel testing - L2 networking using ODL (no east-west security groups), L3 networking uses qrouter/ovs - Overlays: VXLAN, VLAN Colorado 3.0 (Dec/2016) - Enhanced O/S- ODL(L2)-VPP stack (Infra complete: Neutron / GBP Mapper / GBP Renderer / VBD / Honeycomb / VPP) - Enhanced system-level testing - L2 networking using ODL (incl. east-west security groups), L3 networking uses qrouter/ovs - O/S-VPP (Infra: Neutron ML2-VPP / Networking-vpp-agent / VPP) - Automatic Install, Overlays: VLAN Danube 1.0 (March/2017) - Enhanced O/S- ODL(L3)-VPP stack (Infra complete: Neutron / GBP Mapper / GBP Renderer / VBD / Honeycomb / VPP) - L2 and L3 networking using ODL (incl. east-west security groups) BRKSDN-2333 Danube 2.0 (May/2017) - Enhanced O/S- ODL(L3/L2)-VPP stack: HA for OpenStack and ODL (clustering) Euphrates 1.0 (November/2017) - Enhanced O/S- ODL-VPP stack: DVR with VPP (pure L3 for eastwest and northsouth communication); Dynamic VXLANtunnel control via LISP 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 99

96 Tenant network i/f Tenant network i/f DVR with VPP Computenode-0 Controlnode-0 Computenode-1 OpenStack Services Network Control GBP LISP ARPproxy ARPproxy DHCP- Relay DHCP tap L3 VPP FIB Tenant network i/f DHCP- Relay Honeycomb Honeycomb Honeycomb VM 1 VPP L3 FIB NAT VXLAN VXLAN VXLAN NAT L3 FIB VPP vhostuser vhostuser VM 2 External network i/f LISP controlled External network i/f Internet Internet BRKSDN Cisco and/or its affiliates. All rights reserved. Cisco Public 100

97 Create.Compose.Deploy.Test. Iterate Cisco and/or its affiliates. All rights reserved. Cisco Public

98 Cisco Spark How Questions? Use Cisco Spark to communicate with the speaker after the session 1. Find this session in the Cisco Live Mobile App 2. Click Join the Discussion 3. Install Spark or go directly to the space 4. Enter messages/questions in the space cs.co/ciscolivebot#brkdcn Cisco and/or its affiliates. All rights reserved. Cisco Public

99 Please complete your Online Session Evaluations after each session Complete 4 Session Evaluations & the Overall Conference Evaluation (available from Thursday) to receive your Cisco Live T-shirt All surveys can be completed via the Cisco Live Mobile App or the Communication Stations Complete Your Online Session Evaluation Don t forget: Cisco Live sessions will be available for viewing on-demand after the event at Cisco and/or its affiliates. All rights reserved. Cisco Public