OpenStack Summit Sydney: Red Hat talks NFV

In just two weeks, people from around the world will gather in Sydney, Australia for the OpenStack Summit Nov. 6-8. At the three-day conference for IT business leaders, cloud operators and developers covering the open infrastructure landscape, expect to hear and learn about all about modern clouds, and the open technologies being used to build them, including OpenStack, Kubernetes, Docker, Ansible, Ceph, OVS, OpenContrail, OPNFV, and more. Red Hat’s telco team is participating in 10 presentations at the summit. For a closer look at the presentations on network functions virtualization (NFV), read on!

The Road to Virtualization: Highlighting the Unique Challenges Faced by Telcos

Monday, November 6, 11:35am-12:15pm

Sydney Convention and Exhibition Centre – Level 2 – Pyrmont Theatre

In contrast to enterprises and IT virtualization, telcos face a unique set of challenges when entering the network virtualization space. Many of these challenges come down to mapping physical network function (PNF) requirements into virtual network function (VNF) requirements. For instance, the integration with legacy platforms, the integration of existing performance and assurance tools and the skillset refresh of the existing workforce are just a few examples of these challenges.

This presentation will cover:

  • two use cases in development for migrating from PNF to VNF
  • the key challenges telcos face integrating the new network virtualization environments with legacy platforms (for example, achieving 50ms convergence, VRRP, Q-in-Q and LAG)
  • burning questions for operations teams to ensure they are equipped to deal with the new technology

Attendees will learn the real challenges faced by telcos when entering the network virtualization space. The talk will explain how a service provider overcomes those challenges today and what additional work is needed in OpenStack to meet telco requirements in the future.

Presenters:

 

Keeping it Real (Time)

Monday, November 6, 2:15pm-2:25pm

Sydney Convention and Exhibition Centre – Level 2 – Parkside 2 Foyer

Ever get the feeling your cloud is just a bit too…cloudy? The flexibility and resource utilization that the cloud can offer might seem fantastic at first, but things look a little less rosy when you start thinking about the performance impacts the cloud can have on your applications, right?

Never fear: the OpenStack community hears you, and since Juno, projects like Nova have been on a steady march towards adding the features and capabilities that users with NFV and HPC workloads really need from their cloud. The latest of these Nova features builds upon earlier work to finally make real-time workloads in an OpenStack cloud a reality. In this talk, we take a peek under the Nova hood to see how this feature works and how it ties into earlier work in this area. We then demonstrate why this matters and how you can use this feature and others like it in your own applications.

This talk is mainly targeted towards attendees working in the telco/NFV and HPC markets. These attendees are expected to learn about the real-time features recently added to OpenStack Nova. In addition, these features build upon many existing features and we will cover some of these as part of the talk.

By the end of the talk, we would hope that attendees will have vastly expanded their knowledge of OpenStack Nova’s NFV/HPC capabilities, and would be able to use their knowledge of these capabilities in their own applications.

Presenter:

 

Warp-speed Openvswitch: Turbo Charge VNFs to 100Gbps in NextGen SDN/NFV Datacenter

Monday, November 6, 5:50pm-6:00pm

Sydney Convention and Exhibition Centre – Level 2 – Parkside 2 Foyer

Telcos and enterprises are looking for high-performance, secure and scalable network infrastructure while building next-gen datacenters for NFV and SDN deployments. Top requirements are:

  • Scale-up services with analytics: millions of mobile flows; video, voice, data
  • High performance: 25G to 100G per Server
  • Security
  • Optimized resource usage
  • Multi-site deployments: efficient multi-tenancy using SDN overlay
  • Integrated end-to-end solution with no vendor lock-in

In this talk, Red Hat, Nuage Networks and Mellanox will address these challenges in a solution-centric manner, bringing together three key components of NFV and SDN infrastructure: OpenStack, SDN Controller & SmartNICs. We will describe the modifications to many open source components including Linux kernel, OVS & OpenStack (Neutron, Nova & OVS ML2), to enable a vendor-agnostic, open-source and community-driven OVS offload solution that’s radically different, offers blazing high speeds up to 100Gbps/VNF and has no trade-offs.

Attendees will learn:

  • Requirements for next-gen SDN/NFV datacenters: 5G, IoT and mobile applications will demand more from the infrastructure
  • What is OVS Offload to SmartNICs? How does it meet the needs of next-gen SDN/NFV Datacenter? — 25G+ speeds with SR-IOV no CPU overhead. Offload OVS flow rules and actions, traffic classification, QoS marking, encap/decap tunneling, NAT, firewall ACL filtering, TCP connection tracking, DPI and DDoS prevention. If offload fails, fall back to OVS slow-path. Offload millions of flows with firewall rules and connection tracking.
  • Tradeoffs between data path options; SR-IOV, OVS-DPDK and OVS Offload
  • Dive into OVS Offload options: SR-IOV with OVS offload, partial offload with OVS-DPDK, virtio options with live migration
  • Community activities including Linux kernel TC/flower, OVS match and action offload, OpenStack upstream work for Neutron, Nova and ML2 (OS-VIF)
  • Limitations of OVS offload with SmartNICs – Features dependent NIC vendor, bonding within NIC only, limited live migration support with virtio
  • Proof of Concept results of integrated solution with performance analysis of OVS Offload with VLAN and VXLAN

Presenters:

 

ETSI NFV Specs’ Requirements vs OpenStack Reality

Tuesday, November 7, 9:00am-9:40am

Sydney Convention and Exhibition Centre – Level 3 Convention Center – Cockle Bay Room 1

Five years ago, ETSI NFV started generating the industry momentum around Network Functions Virtualization (NFV) in telco. To this day, it remains the forum with the broadest representation of telco service providers, vendors, regulators and other stakeholders in that space.

From an implementer’s perspective, however, the vast amount of ETSI NFV specs with their technology-agnostic formal language can be daunting. One and a half years ago, a small team of valiant souls from within ETSI NFV set out to decode the cloud API specs and document how far OpenStack meets user requirements expressed in those specs.

We had a great session in Boston where we presented an overview of what is done in the gap analysis and invited users/operators to provide feedback on the results, their priorities from the list, and what else they might see missing. As we were able to cover only a small fraction of the gaps in that first session, we believe there is a need to continue the discussion in Sydney.

Attendees can expect to learn more about the discovered gaps, the motivation behind the requested functionality and possibly a design alternative to close the gaps.

Presenters:

 

Turbo charging OpenStack for NFV workloads

Tuesday, November 7, 3:20pm-3:30pm

Sydney Convention and Exhibition Centre – Level 2 – Parkside 2 Foyer

The fundamental business model for communications service providers (CSPs) and telcos is based on providing mission-critical applications to a large pool of subscribers with the least amount of service disruption. There are two key requirements for NFV – performance and high availability. Without being able to achieve very high throughputs that are comparable to what is achievable with purpose-built hardware solutions, this whole business model breaks down. High availability and resiliency is just as important to substantiate the performance throughput for telco clouds.

The journey began with using Linux bridges, virtio and OVS. As demand for higher performance grew in NFV, PCI passthrough, SR-IOV, OVS with DPDK and Vector Packet Processing (VPP) were introduced to meet the demand. We cover each of them in this session with their use cases for NFV workloads (VNFs) while comparing and contrasting the different technologies with details of how to deploy, troubleshoot and operate at scale.

Attendees will learn concepts of various virtual networking technologies required to enable NFV, the advantages and disadvantages of each technique and their application in NFV use cases. They will enhance their knowledge on dataplane performance tuning and learn about best practices for building NFV OpenStack clouds. The attendees will learn to design, configure and deploy OpenStack with the details of configuration parameters as well as troubleshooting tools. This session will go through day 0, day 1 and day 2 aspects of deploying high-performance clouds.

Presenters:

 

We hope to see you in Australia next month, Nov. 6-8, at the Sydney International Convention Centre. We’ll be sharing more details on other telco-related topics in the next week or so, but visit the OpenStack Summit event site for registration details and a preview of the summit agenda.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.