Visit the OpenStack Summit page for the latest news, registration and hotels.

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Networking [clear filter]
Monday, November 3


IPv6 Features in OpenStack Juno
During the Juno development cycle, numerous IPv6 features were merged into Neutron. Join Xu Han Peng, Sean M. Collins, and Robert Li as they discuss the following features available in OpenStack Neutron, Juno release:

  • Provider Networking - Upstream SLAAC Support.

    • Cloud admins create a Provider Network in Neutron, that use an upstream router that advertises RA s - so that instances can use SLAAC to configure their IPv6 networking.

  • RADVD - Support Router Advertisements using RADVD

    • Support for radvd in Neutron, for Neutron routers to announce IPv6 routes and addresses

  • Dnsmasq - DHCPv6 stateless and stateful mode in Dnsmasq

    • Adds support for stateful/stateless DHCPv6 to configure instances

This session will also discuss features that are planned for the upcoming Kilo release, such as:

  • Support Multiple IPv6 Prefixes and Addresses for IPv6 Network

    • Multiple IPv6 prefixes and/or addresses per Neutron port will be supported in OpenStack so a mix of public and private IPv6 addresses can supported.

  • Support IPv6 Prefix Delegation

    • Prefix delegation can be used to automatically configure neutron routers with prefixes so that IPv6 prefixes are obtained and renumbering can be done automatically.

  • Floating IP support for IPv6 Addresses

  • Add support for dual-stack (IPv4 and IPv6) on external gateway

  • Dual-stack arrangement on a router's gateway port can be supported by allowing both an IPv6 and an IPv4-based subnet to be associated with a router.


Sean M. Collins

Senior Software Engineer, Comcast
Sean Collins is a Senior Software Engineer at Comcast and is known for his work on Neutron. Sean has been hard at work introducing IPv6 support for OpenStack, landing his changes in the Juno release.

Robert Li

Technical Leader, Cisco Systems
Robert is a technical lead engineer at Cisco. For the past several years, I have been contributing to cloud infrastructure software development, especially in OpenStack. My current focus is to contributing to the Kubernetes & Istio communities.
avatar for Xu Han Peng

Xu Han Peng

Software Engineer, Subject Matter Expert in OpenStack Networking, IBM

Monday November 3, 2014 11:40 - 12:20
Amphitheatre Bleu


The Open NFV Organization, Neutron and Open Daylight
A new organization has been formed called Open Network Function Virtualization (OP-NFV) under the Linux Foundation.  This new organization seeks to create integration projects that combine upstream code from projects such as Openstack, Opendaylight and the Linux Kernel and combine them with extensive testing, custom configuration and possibly upstream code patches. These things are based on the ETSI framework and requirements. 

This presentaiton will first describe the organization, its motivations and purpose. It will then present several key areas that that organization can collaborate with Openstack and Open Daylight. In the case of the former, the presentation will detail modifications to Neutron and Nova that can be used to better position virtualized network function elements, and how to initially orchestrate them. The next portion of the presentation will focus on how Open Daylight can be used as an SDN controller to complete the orchestration operation, and continue to manage the NFV deployment. 

Finally, the presentation will demonstrate for the viewers, a running example of the aforementioned configuration as a means of showing one possible configuration that meetings the NFV requirments the OP-NFV organization has set out to satisfy.


The presentation will be co-presented by myself and Chris Wright from Red Hat.



avatar for Tom Nadeau

Tom Nadeau

Distinguished Engineer, Brocade
Tom is a Distinguished Engineer at Brocade where he is The Chief Architect of the Open Source projects, and leads teams developing an Open Daylight-based Controller, applications build on that controller and Open Stack projects. Tom also participates in the ONF and IETF, where he... Read More →
avatar for Chris Wright

Chris Wright

CTO, Red Hat
Chris Wright is vice president and chief technology officer (CTO) at Red Hat. Wright leads the Office of the CTO, which is responsible for incubating emerging technologies and developing forward-looking perspectives on innovations such as artificial intelligence, cloud computing... Read More →

Monday November 3, 2014 12:30 - 13:10
Amphitheatre Bleu


Bridges and Tunnels: A Drive Through OpenStack Networking

Neutron provides a rich set of logical constructs to develop a full featured virtual network deployment. In this talk, we will tour a basic Neutron setup and examine the IPv6, distributed routing and load balancing components.  We will close with look back on the new features in Juno and look ahead to the planned work in Kilo.


Mark McClain

Neutron PTL & Senior Principal Architect, Yahoo!
Mark McClain is a Senior Principal Architect at Yahoo! and is the OpenStack Networking Project Technical Lead. Formerly of DreamHost, he has 14 years of software development experience and OpenStack Networking combines two of his favorite interests: networking and Python.
avatar for Salvatore Orlando

Salvatore Orlando

Distinguished Procrastinator, VMware
Salvatore used to be a member of the Neutron core & drivers teams, until he decided he made enough damage and left. He now spends his days picking colors for painting bike sheds and shaving yaks. He has been working in the Openstack networking project since its inception. He actually... Read More →

Monday November 3, 2014 14:30 - 15:10
Amphitheatre Bleu


Architectural Overview of Distributed Virtual Routers in OpenStack Neutron
Since the Diablo days, Nova has had a multi-host feature that offers attractive network
performance, scalability and availability characteristics.  This has long been seen as a gap
in Neutron, where routing and floating IPs have been provided by a centralized component.

With the Juno release, Neutron introduces a Distributed Virtual Router (DVR) feature.   DVR brings
improvements in network performance, scalability and availability by distributing the routing and
floating IP functions.

In this session, members of the DVR upstream team behind the DVR implementation will discuss
the design and architecture of the DVR feature.  We will provide insight into key objectives and
design decisions that lead to the Juno implementation.  We will explore the differences between
a deployment using centralized routers and one using DVR.  We will also discuss limitations of the
current DVR implementation, and take a look forward at future related work.

Attendees will leave this session with a clear understanding of how Distributed Virtual Router ties into the OpenStack Neutron framework to support the Openvswitch based deployments that addresses the scalability, performance and availability issues.


Rajeev Grover

Systems Software Engineer, HP, HP
Rajeev is a systems software engineer at HP Networking. His interests are in Networking, Servers and Virtualization. He has a Masters in Computer Science with over 20 years of industry experience. He is an active member and follower of OpenStack Neutron. 

Jack McCann

Distinguished Technologist, Hewlett Packard Enterprise
Jack McCann is an architect in Helion Openstack engineering.

Vivekanadan Narashimhan

Systems Software Engineer, Hewlett Packard Company
Expertise in Networking domain and active contributor/follower of Openstack Neutron

Michael Smith

Michael has been working in software engineering for over 16 years and has been most recently working on OpenStack networking agents for HP Networking.
avatar for Swaminathan Vasudevan

Swaminathan Vasudevan

System Software Engineer, Hewlett-Packard Company (HP)
Swaminathan Vasudevan is a Systems Software Engineer at HP where he currently works on OpenStack Neutron and OpenDaylight. He is an active technical Contributor of OpenStack since 2013. Swaminathan Vasudevan has been developing code for more than 18 years with expertise in linux... Read More →

Monday November 3, 2014 15:20 - 16:00
Amphitheatre Bleu


Group Based Policy Extension for Networking
The Group Based Policy (GBP) extension introduces a declarative policy driven framework for networking in OpenStack. The GBP abstractions allow application administrators to express their networking requirements using group and policy abstractions, with the specifics of policy enforcement and implementation left to the underlying policy driver. This facilitates clear separation of concerns between the application and the infrastructure administrator.

Over the past two release cycles, the GBP model has been incubated in Neutron as an extension. A new sub group has worked on defining this extension and  the proposed specification has been approved for implementation in the Juno release cycle. In this talk, we first discuss the GBP extension API and then present the reference implementation for it. In particular, we show how a new Service plugin is designed and developed in order to support this extension through a framework of configurable policy drivers similar to that used in the ML2 plugin. We will showcase the latest working version of the code, and provide an end-to-end demonstration of the features. We will also present several vendor and open source policy drivers that are being developed to support this new extension.


The work in Juno implements a subset of a more richer model that is intended for development in Kilo. We will provide a roadmap for the upcoming features including the integration with the advanced services (L4-7) framework. We will also discuss how the new extension can be utilized by other OpenStack projects.


avatar for Mohammad Banikazemi

Mohammad Banikazemi

Research Staff Member, IBM Research
Mohammad is a research staff member at the IBM T.J. Watson Research Center. His research interests include cloud computing and software-defined networking. He is a senior member of the ACM and the IEEE and an active contributior to Neutron. Mohammad lives with his family in NYC.

Sumit Naiksatam

Principal Engineer, Cisco Systems
Sumit Naiksatam is a Principal Engineer at Cisco Systems. He has been involved with Neutron as a core developer since it's inception. He is actively involved in driving and implementing the Group Policy model. He is also driving the efforts around Group-based Policy, integration of... Read More →

Hemanth Ravi

Hemanth Ravi works as CTO at One Convergence and is resonsible for the technology direction and architecture of their solution providing network virtualization and deployment of network services. His experience includes working as an architect of products in data networking such as... Read More →

Stephen Wong

Software Architect, FutureWei Technologies, Inc.
Stephen works for FutureWei Technologies, Inc. and has had nearly ten years of open source software project contributions. He is currently the PTL (Project Team Lead) of OPNFV Clover, a project which utilizes CNCF and related projects for NFV use cases. Previously he was core reviewer... Read More →

Monday November 3, 2014 16:20 - 17:00
Amphitheatre Bleu


Load Balancing as a Service v2.0 - Juno and Beyond
Load balancing as a service has been one of the critical features asked for by cloud tenant.

For the Juno release cloud providers such as Rackspace, HP, etc. have partnered with the community and load banalcer vendors such as Radware and A10 to redefine the load balancing as a service APIs to address the accomulated experience and tenant needs.

In this session we will discuss the process leading to the new API, What is delivered for Juno and what is the road map going forward.

avatar for Samuel Bercovici

Samuel Bercovici

Director, Radware
Samuel Bercovici is director at Radware in charge of integration Radwares load balancer and security services with cloud systems. He was involved at the specification and design of Load Balancing as a Service for Folsom and Grizzly as well as the new API for Juno. During the different... Read More →
avatar for Brandon Logan

Brandon Logan

Software Developer, Rackspace
Core reviewer for neutron-lbaas and octavia projects.  
avatar for Phillip Toohill

Phillip Toohill

Software Engineer, Rackspace
Software developer for the Rackspace team committed to expanding capabilities and performance of OpenStack Networking services with primary focus on Load Balancers (LBaaS).
avatar for Doug Wiegley

Doug Wiegley

Director of SW Engineering, A10 Networks
Doug has worked at companies ranging from 2 person startups to a Fortune 100 companies. He has designed and implemented cloud based mobile application solutions, email applications and networking software. His wide range of work experience allows him to easily switch from high level... Read More →

Monday November 3, 2014 17:10 - 17:50
Amphitheatre Bleu
Tuesday, November 4


La Vie d un Paquet
OpenStack deployments for public or private clouds require strategies for building production-ready networks. Today s deployments require performance, resilience, and high availability to be considered truly production-ready.

In this session, we deep dive into the process of sending a data packet across an OpenStack environment through a network overlay. In a distributed architecture, network functions are applied in a single virtual hop. By applying network services right at the initiating host, the network is free from unnecessary clogging and bottlenecks by avoiding additional hops. 

A distributed architecture implements logical constructs that are used to build networks without a single point of failure, all while adding network functionality in a highly-scalable manner. Packets reach their destination more efficiently with the single virtual hop.

After this session, the audience will understand how distributed architectures allow efficient networking with routing decisions and network services applied at the edge. Also, the audience will understand how it s easier to scale clouds when the network intelligence is distributed.

avatar for Cynthia Thomas

Cynthia Thomas

Networking Specialist, Google
Cynthia Thomas (@_techcet_) is a Networking Specialist at Google Cloud. Her background includes working with open source cloud & networking solutions. Since 2015, she has been working on Docker and Kubernetes with CNI plugins, currently through the open source project Cilium (www.cilium.io... Read More →

Tuesday November 4, 2014 11:15 - 11:55
Amphitheatre Bleu


OVX: Virtual Software Defined Networks
avatar for Ali Al-Shabibi

Ali Al-Shabibi

Network Software Developer, Open Networking Laboratory
Ali Al-Shabibi is the lead engineer and maintainer of OpenVirteX, a network hypervisor and core developer of ONOS at the Open Networking Laboratory. Previously, he was a post-doc at Stanford University researching OpenFlow and SDNs in Nick McKeown's group. He received his Ph.D from... Read More →

Tuesday November 4, 2014 12:05 - 12:45
Amphitheatre Bleu


Akanda: Layer 3+ Virtual Networking Services for OpenStack
DreamHost has been working on our OpenStack Public Cloud, DreamCompute, for several years. At the onset of the project, we set out with an aggressive set of requirements for our networking functionality, including L2 tenant isolation, IPv6 support from the ground up, and complete support for the then emerging OpenStack Neutron APIs. Our search ended with the realization that there was a gap in OpenStack SDN for L3+ services. Thus, the Akanda project was born.

Akanda is an open source suite of software, services, orchestration, and tools for providing L3+ services in OpenStack. It builds on top of OpenBSD, Packet Filter, and OpenStack Neutron, and is used in production to power DreamCompute's networking capabilities. Using Akanda, an OpenStack provider can provide tenants with a rich, powerful set of L3+ services, including routing, port forwarding, firewalling, and more.

This talk will give an introduction to the Akanda project, review the DreamCompute use case, and illustrate how Akanda works under the hood. In addition, we'll discuss future capabilities, operational challenges and tips, and more.

avatar for Jonathan LaCour

Jonathan LaCour

VP, Cloud, DreamHost
Jonathan has been programming from a young age, finessing his way through many programming languages. After cutting his teeth with several smaller jobs and personal projects, Jonathan joined VertiSoft Corp. in 1997, catapulting him into the culture of "software as a service" well... Read More →
avatar for Ryan Petrello

Ryan Petrello

Senior Cloud Developer, DreamHost, DreamHost
Ryan has been involved with OpenStack since he joined DreamHost in 2012, where he has worked on DreamCompute. Ryan has contributed to a large number of OpenStack projects, and is the primary maintainer for the Pecan web framework that powers many OpenStack APIs. Ryan is also heavily... Read More →

Tuesday November 4, 2014 14:00 - 14:40
Amphitheatre Bleu


Neutron Network Node High Availability
Today, you can configure multiple network nodes to host DHCP and L3 agents. In the Icehouse release, load sharing is accomplished by scheduling virtual routers on L3 agents. However, if a network node or L3 agent goes down, all routers scheduled on that node will go down and connectivity will be lost. The Juno release will introduce L3 high availability as well as distributed routing.

L3 high availability schedules routers on more than a single network node. Stand-by routers use keepalive (And the VRRP protocol internally) to monitor the active router and pick up the slack in case of a failure. We'll explain what is VRRP and how will the feature affect the network node, as well as give a demonstration.

Distributed routing, or DVR, moves routing to the compute nodes, while keeping SNAT on the network nodes.

We'll talk about DHCP HA, go into a conceptual overview of both L3 HA and DVR, give demonstrations, and explain how to integrate them together to get the best of both worlds.


Sylvain Afchain

Software Engineer, eNovance
Sylvain Afchain is a Senior Developer at eNovance. He has 15 years of software development experience. He has been involved on Openstack since two releases. He worked mainly on Neutron and on Network projects.
avatar for Assaf Muller

Assaf Muller

Manager of OpenStack Neutron Engineering, Red Hat
twitter.com/assafmuller Manager of OpenStack Neutron Engineering at Red Hat. Neutron core team member and responsible for the project’s testing. Assaf works out of sunny Boston, USA.

Tuesday November 4, 2014 14:50 - 15:30
Amphitheatre Bleu


Can You Trust Neutron? A Tour of Scalability and Reliability Improvements from Havana to Juno
It is a widely known fact that Neutron has been for a long time under the spotlight for being the "naughty" openstack integrated project, constituting, to an extent, Openstack's weak link.

The network is a crucial component, and Neutron provides a network virtualization framework which enables layer-2 tenant isolation, VIF level firewall with security group as well as layer-3 routing, as well as several L4/L7 services.

This talk is aimed at discussing the evolution of this framework from havana to Juno.

However, rather than focusing on aspects such as features and capabilities, this talk will rather focus on aspects which probably are a lot more important to deployers, such as reliability and scalability.

The talk agenda can be summarized as follows:

- An overview of the issues affecting Openstack neutron at the Havana release

- A discussion of the improvements made to Neutron, and in some cases to Nova as well

- A quantitative measuring of the improvements occurred between Havana and Juno in terms of performance, and scalability. This will be achieved comparing Neutron's "reference" Open source solution (ML2 with OVS) between these two release, as well as comparing it against another plugin which will be used as a reference to measure the progress of Neutron's network virtualization framework.

avatar for Salvatore Orlando

Salvatore Orlando

Distinguished Procrastinator, VMware
Salvatore used to be a member of the Neutron core & drivers teams, until he decided he made enough damage and left. He now spends his days picking colors for painting bike sheds and shaving yaks. He has been working in the Openstack networking project since its inception. He actually... Read More →
avatar for Aaron Rosen

Aaron Rosen

Developer/Neutron Core
  Currently hacking on OpenStack for Nicira/VMware and is a member of the core neutron and core congress team. Lately, has been hacking on Congress to help provide a governance and compliance layer for dynamic infrastructure. When not working on that he works on the Networking (Neutron... Read More →

Tuesday November 4, 2014 15:40 - 16:20
Amphitheatre Bleu


Designate: An Overview of OpenStack DNSaaS
Designate, a recently incubated Project, intends to solve the DNS needs of OpenStack by providing a RESTful API for managing DNS data with a variety of databases on many different backends. Additionally, Designate will also provide a service for integrating with Nova and Neutron. This talk will cover:

  • Why Openstack and operators need DNSaaS.

  • The Designate Architecture.

  • The Designate API.

  • New Features Since the Last Summit, including MiniDNS, Server Pools, the Version 2 API.

  • The Next Steps For Designate, including the Version 2 API, additions to Server Pools, and MinDNS.

Come see who is involved, and how easy it is to get involved in this exciting project.

This talk is a general overview. It will cover the problem statement, architecture and general features.

avatar for Graham Hayes

Graham Hayes

Senior Software Engineer, HP Cloud Services
Graham is the newly elected PTL for Designate in Mitaka, and lead the design work on the latest major feature for Designate - Server Pools. Graham works as part of the DNSaaS team in HP Helion. As part of this team he is responsible for operating a publicly accessible deployment... Read More →

Endre Karlson

Software Engineer @ HP Cloud

Vinod Mangalpally

Software Developer @ Rackspace
avatar for Tim  Simmons

Tim Simmons

Software Developer, Rackspace
Tim is a Software Developer at Rackspace on the Cloud DNS team, and a member of Designate core. He has been working on Designate for over two years, and is working on operating Designate at scale for Rackspace.

Tuesday November 4, 2014 16:40 - 17:20
Amphitheatre Bleu


Managing Open vSwitch across a large heterogenous fleet
Open vSwitch (OVS) is one of the more popular ways to provide VM connectivity in OpenStack. Rackspace has been using Open vSwitch in production since late 2011. In this session, we will detail the challenges faced with managing and upgrading Open vSwitch across a large heterogenous fleet. Finally, we will share some of the tools we have created to monitor OVS availability and performance.

Specific topics covered will include:


  • Why upgrade OVS?

  • Measuring OVS

  • Minimizing downtime with upgrades

  • Bridge fail modes

  • Kernel module gotchas

  • Monitoring OVS

avatar for Andy Hill

Andy Hill

Systems Engineer, Rackspace, Rackspace Hosting

Trey Morris

Software Developer, Rackspace
avatar for Joel Preas

Joel Preas

Public Cloud Systems Engineer
A cog in the machine at the Rackspace public cloud.

Tuesday November 4, 2014 17:30 - 18:10
Amphitheatre Bleu
Wednesday, November 5


Migrating Production Workloads from OVS to Linux Bridge w/ ML2
While OpenVswitch has proven incredibly useful for environments with a need for rapidly changing networks, it has some shortcomings in more compact environments with a less robust or redundant infrastructure. In order to provide a more stable environment for our customers, we have chosen to use the ML2 plugin with Linux Bridge.

We'll discuss how we were able to migrate private cloud environments from OpenVswitch using the OVS plugin to the Linux Bridge plugin under ML2 without having to redefine networks or unplug or re-IP instances. We'll demonstrate configuration file and database changes, procedures and best practices we found during the migration. Finally, we will discuss the pros/cons and capabilities of each product and make users aware of any caveats that they may expect with this particular migration.


James Denton (Openstack Network Architect - Rackspace)

Kevin Stevens (Openstack Engineer - Rackspace)

avatar for James Denton

James Denton

Principal Architect, Rackspace
James Denton is a Principal Network Architect for Rackspace, and has been involved with OpenStack on the Rackspace Private Cloud team since 2012. He is the author of "Learning OpenStack Networking (Neutron)" released in 2014, and is currently working on a second edition of the same... Read More →
avatar for Kevin Stevens

Kevin Stevens

Openstack Engineer
Kevin Stevens is an Openstack Engineer on the Private Cloud team at Rackspace. He has operated/built Openstack clouds for customers with a range of technical needs for more than 3 years. Kevin is intimately familiar with Nova-Network and Quantum/Neutron and what it takes to be successful... Read More →

Wednesday November 5, 2014 09:00 - 09:40
Amphitheatre Bleu


Neutron Roundtable: Overlay or Full SDN?

Neutron offers multiple ways to implement networking. It's not only a matter of vendor choice, but also a choice of networking models. Should the tenants of your cloud be allowed to place requests that would directly modify the configuration of your hardware, or would you like them confined in virtual land? What are the limits of each models and can they be combined? Why would you need access to BGP/OSPF layers from Neutron? What about VPNs or MPLS?

In this roundtable we will ask 5 OpenStack Networking experts to prepare a 5 min position statement on which model they prefer and for what purposes, then we will open the floor to a debate within the group and with the public.

avatar for Nick (Nicolas) Barcet

Nick (Nicolas) Barcet

Director of Product Management - OpenStack, Red Hat
Nick is the Director of Product Management for OpenStack at Red Hat.  Nick joined Red Hat in June 2014, as part of the acquisition of eNovance, where he was VP of Products. Prior to that role, Nick was Ubuntu Server and Cloud Product manager at Canonical, where he participated... Read More →

avatar for Mike Cohen

Mike Cohen

Cisco, Director of Product Management
Mike Cohen is Director of Product Management at Cisco Systems where leads open source development for the Insieme business unit. Mike began his career as an early engineer on VMware's hypervisor team and subsequently worked in infrastructure product management on Google and Big Switch... Read More →
avatar for Dan Mihai Dumitriu

Dan Mihai Dumitriu

CEO and Co-Founder, Midokura
Dan is responsible for the technical innovation and development of designing, building and operating Midokura technology. Prior to founding Midokura, Dan served as Chief Architect at Ballista Securities, a New York City ATS offering an electronic block trading system for options... Read More →
avatar for David Lenrow

David Lenrow

Distinguished Architect, HP
David Lenrow was trained as a computer scientist and has spent more than 20 years driving innovation in digital technology with an emphasis on networks, storage and media.  His career spans multiple roles from individual contributor to executive across all major functional areas... Read More →
avatar for Pedro Marques

Pedro Marques

Software Engineer, Juniper Networks
Software architect focusing on networking platforms. Experience in control plane (routing and switching), data plane, ASIC definition and infrastructure. Focus on building systems from initial concept to product stage.
avatar for Chris Wright

Chris Wright

CTO, Red Hat
Chris Wright is vice president and chief technology officer (CTO) at Red Hat. Wright leads the Office of the CTO, which is responsible for incubating emerging technologies and developing forward-looking perspectives on innovations such as artificial intelligence, cloud computing... Read More →

Wednesday November 5, 2014 09:50 - 10:30
Amphitheatre Bleu