April 3-5 | San Jose, California
View More Details  & Register Here

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

Cloud Native Networking (for Carriers and Enterprise) [clear filter]
Wednesday, April 3

3:10pm PDT

Network Service Mesh: An Introduction - Ed Warnicke, Cisco Systems & Frederick Kautz, Doc.ai
Network Service Mesh (NSM) is a novel approach solving L2/L3 network use cases in Kubernetes that are tricky to address with the existing Kubernetes Network Model. Inspired by Istio, Network Service Mesh maps the concept of a service mesh to L2/L3 payloads.

Network Service Mesh Use Cases such as:

* Connecting Pods to:
-VPN Gateways
-Non-Kubernetes virtual bridge domains
-The *right* physical NIC or SR-IOV VF for their needs
* Cloud-native NFV use cases

Network Service Mesh controls the L2/L3 data planes to deliver these types of use cases. Network Service Mesh enables users to express the context of their network needs in a Cloud Native manner, rather than manually stringing together disjoint interfaces, IPAM, and subnets.

Finally, we discuss how audience members can get involved and help drive the direction and development of NSM.

avatar for Frederick Kautz

Frederick Kautz

Head of Edge Infrastructure, Doc.ai
Frederick Kautz is Head of Edge Infrastructure at Doc.ai. He was previously a Principal Software Engineer in the Office of Technology at Red Hat where he focused on improving the overall state of container networking and container+SDN integration. Frederick is an active contributor... Read More →
avatar for Ed Warnicke

Ed Warnicke

Distinguished Engineer, Cisco
Ed Warnicke is a Distinguished Engineer at Cisco Systems. He has been working for nearly two decades in many areas of networking and Open Source.  Ed is currently the FD.io TSC Chair, and a founder of and active contributor to the Network Service Mesh project. Ed has a masters in... Read More →

Wednesday April 3, 2019 3:10pm - 3:40pm PDT

3:50pm PDT

Works on My Machine: How to Validate Performance of VNFs and CNFs in a Reproducible Testbed - W. Watson, Vulk Cooperative & Denver Williams, cncf.ci
Are you really testing your network functions? What about the platform it runs on and how the platform was configured for the network+hardware underneath? Come learn how the CNF Testbed demonstrates reproducible test results on bare metal. This session will cover provisioning the network and hardware, setting up the platform (eg. Kubernetes, OpenStack) and deploying/running test scenarios on the clusters running on bare metal.

avatar for W. Watson

W. Watson

Principal Developer, Vulk Cooperative
W. Watson has been professionally developing software for 25 years. He has spent the numerous years studying game theory and other business expertise in pursuit of the perfect organizational structure for software co-operatives. He also founded the Austin Software Cooperatives meetup... Read More →

Denver Williams

Project Co-Lead, cncf.ci, Vulk Coop & CNCF

Wednesday April 3, 2019 3:50pm - 4:20pm PDT

4:30pm PDT

Square Peg in a Round Hole - Observations Made on Experimenting ONAP Deployment with Legacy Non-cloud Native Ecosystem - Viswanath Kumar Skand Priya, Verizon
This session summaries our observations when we experimented deploying ONAP with legacy current network functions namely NFV-Os, S-VNFMS & OSS Systems. The idea is to share our learnings & identify some of the gaps in ONAP to solve the Brownfield deployment equation.




Wednesday April 3, 2019 4:30pm - 5:00pm PDT

5:10pm PDT

Packet Walk(s) in Kubernetes - Don Jayakody, Big Switch
Have you ever wondered how the packets are getting forwarded inside Kubernetes with different CNI plugins? Whether it’s getting encapsulated? How are ARPs getting resolved? How the packets flow outside from your K8S cluster etc.

When it comes to networking inside Kubernetes, selecting the correct networking solution may be one of the most important decisions you may face when trying to run a Kubernetes cluster in production. Therefore it's beneficial to have a good understanding of different options out there and most importantly how these networking options are different from each other.

In this presentation, I’ll go over packet by packet level details of how the network plumbing is happening with different CNI plugins including, Flannel, Calico & Cilium. I'll cover different packet flows and try to draw different examples so that we all can understand where it's different.

avatar for Don Jayakody

Don Jayakody

Sr. Technical Marketing Engineer, Big Switch
Don Jayakody is a Senior Technical Marketing Engineer at Big Switch Networks. His experience in the networking industry varies from access provider networks, storage area networks to network orchestration systems. He is currently focusing on various deployment methodologies in container... Read More →

Wednesday April 3, 2019 5:10pm - 5:40pm PDT
Thursday, April 4

11:10am PDT

Cloud Native Network Tracing in an Istio Service Mesh Using OPNFV Clover's Clovisor - Stephen Wong, FutureWei Technologies, Inc.
Clovisor, from OPNFV Clover project, is a cloud independent and CNI-plugin agnostic network tracing module. It augments tracing data from Istio service mesh via providing raw network traces on a per session/transaction basis, and implementing a golang plugin infra to support multiple network protocols.

Clovisor integrates tightly with Kubernetes and Istio, and utilizes IOVisor/goBPF to tap into node and k8s pod interfaces to trace packets, which minimizes impact on kernel packet forwarding. It uses OpenTracing APIs to send to Jaeger tracer its network traces, and features a sample analytic engine to correlate traces from Istio and Clovisor. These resulting consolidated traces are particularly useful for deployments in:
o Hybrid cloud
o Edge computing
o Multi-site deployment

This presentation will give detail introduction to Clovisor, its architecture and design, and will provide a demo


Stephen Wong

Stephen Wong has nearly ten years of experiences contributing to various open source communities. He is currently the PTL (Project Team Lead) of OPNFV Clover, a project which utilizes cloud native related projects for NFV (network function virtualization) use cases. Previously he... Read More →

Thursday April 4, 2019 11:10am - 11:40am PDT

11:50am PDT

Panel Discussion: Towards Kubernetes as the Network Service Orchestrator - Heather Kirksey, The Linux Foundation; Ramki Krishnan, VMware; Ed Warnicke, Cisco Systems; Jeffrey Saelens, Charter Communications; and Nikolay Nikolaev, VMware
In the current landscape, network services are defined by a slow-moving standards ecosystem (TOSCA etc.) with vendor proprietary extensions. With the advent of Edge Computing and 5G, dynamism of Network Services is a critical element in achieving end-to-end service agility for applications and networks for enterprises, network operators and beyond. While transition from PNFs->VNFs->CNFs is an important element of this transformation, delivering a dynamic network service using an industry standard API is a key gap to address in a disaggregated landscape.

This panel discusses how the open source effort Network Service Mesh (NSM) uses the extensibility of Kubernetes API to deliver dynamic network services across Distributed DC deployments by seamlessly interconnecting PNFs, VNFs and CNFs at Layer 2/3. This panel also elaborates on the current state of NSM and future architectural plans.

avatar for Ramki Krishnan

Ramki Krishnan

Lead Technologist, Open Source, VMware
Ramki, with 20+ years of industry experience, has a deep understanding of various technologies and strong business acumen to lead and transform innovation into customer-winning products. Currently, at VMware, he is responsible for Telco/Enterprise open source technology vision, strategy... Read More →
avatar for Ed Warnicke

Ed Warnicke

Distinguished Engineer, Cisco
Ed Warnicke is a Distinguished Engineer at Cisco Systems. He has been working for nearly two decades in many areas of networking and Open Source.  Ed is currently the FD.io TSC Chair, and a founder of and active contributor to the Network Service Mesh project. Ed has a masters in... Read More →
avatar for Nikolay Nikolaev

Nikolay Nikolaev

Technical Lead, Kong Inc.
Nikolay Nikolaev is a Technical Lead in the Office of the CTO at Kong Inc. where he works on the Kuma Envoy Control Plane. For the last 17 years, he has been implementing networking software ranging from hardware boxes to powerful server applications and virtualized data planes. He... Read More →

Jeffrey Saelens

Principal Cloud Architect, Charter Communications
Jeffrey Saelens is a Principal Cloud Architect in Charter Communications’ Advanced Engineering commercial business unit. Starting his career in the US Army, Jeffrey was a Green Beret focused on communications and systems engineering. After leaving the military, he dove into the... Read More →
avatar for Heather Kirksey

Heather Kirksey

Vice President of NFV, The Linux Foundation
Heather Kirksey works with the community to advance the adoption and implementation of open source NFV platforms.Before joining The Linux Foundation, she led strategic technology alliances for MongoDB. Earlier in her career she held various leadership positions in the telecom industry... Read More →

Thursday April 4, 2019 11:50am - 12:20pm PDT

1:50pm PDT

A Comparison of Service Mesh Solutions: Looking at Istio, Linkerd and Consul Connect - Syed Ahmed, CloudOps
As you move from a monolithic to microservices-based architecture, you'll run into a variety of challenges that can be addressed through the adoption and use of a service mesh. A service mesh is a dedicated infrastructure layer for managing service-to-service communication to make it visible, manageable, and controlled. In this presentation, we'll compare and contrast the leading service mesh solutions - Istio, Linkerd and Consul Connect - with emphases on their use in Kubernetes (container) and networking environments.


Syed Ahmed

Cloud Software Architect, CloudOps
Syed Ahmed is a cloud software architect at CloudOps. With extensive knowledge throughout both hardware and software stacks, he adds a unique perspective to solving integration and orchestration challenges. Syed is an avid open source contributor, and a Committer and PMC in the Apache... Read More →

Thursday April 4, 2019 1:50pm - 2:20pm PDT

2:30pm PDT

Identity-based Cross-cluster Fabrics - Challenges & Rewards - Igor Tarasenko, Bayware
Cloud-native infrastructures and platforms have radically improved application deployment with one major change still to come - replacing IP addresses with Identity - as defined by applications - for cross-cluster service-to-service communication.
Challenges include deriving application intent and identity directly from the application manifest, enabling cross-domain networking based only on identity, making flows directly programmable as code, and achieving authorization and authentication of identity as well as flow telemetry and code updates that are virtually instant. Rewards include making NetOps a reality, ie bringing network policy as code into CI/CD pipelines, creating networks that dynamically respond to applications and conditions, giving applications full observability to flow-level telemetry. Also, this enables compute-in-network functions for advanced AI and IOT applications.


Igor Tarasenko

CTO, Bayware
Igor is co-founder and CTO of Bayware in San Francisco. He is an innovative IT and telecom technologist with 20+ years experience. Igor launched his 20-year telecom career at Ukrtelecom where he held CTO and CIO positions developing and deploying services and applications for 10M... Read More →

Thursday April 4, 2019 2:30pm - 3:00pm PDT

3:40pm PDT

Running TungstenFabric at Scale - Mick McCarthy & David O'Brien, Workday
TungstenFabric (TF) is an open source, multi-cloud, multi-stack software-defined networking (SDN) platform. It works with various orchestration stacks like OpenStack, Kubernetes, VMWare and Mesos. It has been gaining traction in the Telecom as well as Enterprise private/public clouds.

TungstenFabric has a large and rapidly growing footprint at Workday for powering networking for the OpenStack based Workday Private Cloud. Running TF at Scale involves optimizing various different components of the TF stack, external dependencies and parts of the underlying physical network. In this talk, Mick & David will discuss what scale looks like from a Workday Private Cloud perspective and go through Workday's journey - challenges, optimizations and lessons learnt of running TF at scale.

avatar for Mick McCarthy

Mick McCarthy

Software Engineer, Workday
avatar for David O’Brien

David O’Brien

Software Development Engineer, Workday

Thursday April 4, 2019 3:40pm - 4:10pm PDT

4:30pm PDT

Lightning Talk: Secondary Network Interfaces for Containers, Its Types and Use-cases - Tomofumi Hayashi, Red Hat
The last year, 2018, Kubernetes Network Plumbing Group (NPWG) successfully published the first de-facto standard document for network attachments, with reference CNI plugin, Multus-CNI. It makes us to having two or more interface to the pods/containers in Kubernetes. But it is not the goal, just a start of the journey of advanced container networking. "Which interface is good for secondary network?", that is the first step. This talk will survey over several CNI interface plugin and CNI IPAM plugin, like ipvlan, macvlan, DHCP and static and will show best practices for several use-cases.

avatar for Tomofumi Hayashi

Tomofumi Hayashi

Principal Software Engineer, Red Hat
Tomofumi Hayashi is a Senior Software Engineer at Red Hat. He works on SDN/NFV development with NFV partners and with open source communities. He focusing on NFVI fault management in OPNFV/Barometer and advanced container networking in Kubernetes networking plumbing working group... Read More →

Thursday April 4, 2019 4:30pm - 4:40pm PDT

5:00pm PDT

Using Kubernetes to Build a Microservices Cloud Native Network OS to Bring CI/CD to Networking - Glenn Sullivan, SnapRoute
The adoption of CI/CD by DevOps-minded organizations has led to more flexible, secure and agile infrastructure with new features and functionality being constantly deployed and improved upon. With these architectures in place, the compute and storage pieces of the environment have benefited from the rapid deployment and constant improvement that CI/CD mandates. Built upon legacy, monolithic architectures — most Network Operating Systems (NOS), are not able to take advantage of CI/CD philosophies. What we built with Kubernetes is a new NOS architecture that is built upon core cloud-native principles of microservices and containerization to enable CI/CD for networking.

This talk gives an overview of adopting Cloud Native principles to bring CI/CD principles to your network. Session attendees will get key take-aways and practical advice on how to implement DevOps in their network.

avatar for Glenn Sullivan

Glenn Sullivan

Co-founder, SnapRoute
Glenn Sullivan co-founded SnapRoute in August 2015, where he is responsible for product direction by advocating for the end-user. Glenn’s experience as an operator of hyperscale data center networks, along with a strong background of technical support gives him a unique skill set... Read More →

Thursday April 4, 2019 5:00pm - 5:30pm PDT
Friday, April 5

11:10am PDT

A Step-by-Step Guide to Building a True Composable Architecture for 5G/Edge - Prem Sankar Gopannan, Lumina Networks
The next generation infrastructure needs composable architecture to realize the needs of 5G/Edge requirements. This session provides an overview of the framework and Opensource tools to build a composable architecture. The session will touch upon Radio (SDN-R), Transport(Network Slicing) and Multicloud management and orchestration solutions (ONAP) the key building blocks of 5G and how cloudnative helps those areas. The session will also provide deep insights into Kubernetes Custom Resource (CRD) models and how it helps in defining a true cloud native architecture.


Prem Sankar Gopannan

Director of Engineering, Lumina Networks
Prem Sankar Gopannan brings two decades of experience to bear as Lumina Networks’ Director of Engineering. At Lumina, he leads 5G product strategy where he leverages his open source expertise. Previous to his role at Lumina, Prem managed SDN product strategy focusing on architecture... Read More →

Friday April 5, 2019 11:10am - 11:40am PDT

11:50am PDT

OpenDaylight, Scale and Containers - Luis Gomez & Anil Vishnoi, Lumina Networks
In this session, speakers will share their experiences from the journey of solving the scalability concern for OpenDaylight using the containers and Cloud Native architecture. Speakers will talk about few important topics like, how they leveraged containers to enable horizontal scalability for OpenDaylight, challenges they encountered in molding OpenDaylight in a Cloud Native architecture, benefits of deploying OpenDaylight in Cloud Native way, new challenges encountered with this approach and finally sharing their thoughts on an interesting question - is OpenDaylight ready for Cloud Native world? Speakers will also share details and best practices about, communication between applications, data persistence, service outage time optimization, security, monitoring and troubleshooting.


Anil Vishnoi

Principal Software Engineer, Lumina Networks
Anil Vishnoi is a Principal Software Engineer for Lumina Networks. Prior to his role with Lumina, Anil has over ten years of experience in data center networking and software defined networking from companies including Brocade, IBM, ISL and Symantec. Anil has six years of open source... Read More →

Luis Gomez

Software Test Engineer, Lumina Networks
Luis Gomez is a seasoned industry engineer with over five years of open source leadership including roles with the technical steering committee, board of directors, and integration groups and is current PTL of distribution and committer for test, builder, autorelease and docs projects... Read More →

Friday April 5, 2019 11:50am - 12:20pm PDT

1:50pm PDT

Accelerated Container Networking using DPDK - Sujata Tibrewala & Muthurajan Jayakumar, Intel Corporation
DPDK leverages the underlying modern hardware architecture to optimize packet processing on Software. Kubernetes as an orchestration system for containers focuses on service velocity, simplicity and scalability. One aspect of that simplicity, container networking by default has only one network interface. However one networking interface is not enough to implement network functions in containers. With the demand of 5G and edge it is becoming more and more critical to implement Container Network Functions. DPDK plug-ins are vital in filling this gap in containers using Conatiner Network Interface, CNI. In this presentation, we will learn how the DPDK plug-ins help in features like Node Feature Discovery, Support for CPU core pinning for K8 pods, Support for Service Chaining aka East West traffic , SR-IOV and control of VNF level – with 100s or 1000s of VNFs.

avatar for Muthurajan Jayakumar

Muthurajan Jayakumar

Platform Application Engineer, Intel Corporation
M Jay has worked with the DPDK team since 2009. M Jay has authored DPDK Cookbook ebook https://software.intel.com/en-us/article/get-the-dpdk-cookbook M Jay joined Intel in 1991 and has worked in various roles and divisions with Intel, including a roles as a 64 bit CPU front side... Read More →
avatar for Sujata Tibrewala

Sujata Tibrewala

Network Technology Evangelist, Intel
Sujata Tibrewala is an Intel community development manager and technology evangelist who defines programs and training events to ensure that the network developer ecosystem works together toward a common goal: to drive SDN/NFV adoption in the industry using open source ingredients... Read More →

Friday April 5, 2019 1:50pm - 2:20pm PDT

2:30pm PDT

Next-Generation NFV Orchestration - Tal Liron, Red Hat
Kubernetes has optimized our ability to deliver applications to the cloud. But what about network service workloads? Efforts to create next-gen "CNFs" notwithstanding, it's possible to support the existing "V" in NFV on Kubernetes right now using technologies like KubeVirt. But is that all we need? ONAP already has some support for deploying VNFs on Kubernetes, but in this presentation we'll look at how Kubernetes itself can take over most of our orchestration tasks.

We'll do this by a gap analysis: What components do we need to add to Kubernetes in order turn into a virtual network service cloud? We'll see that we're much closer to achieving this goal than you might think.

avatar for Tal Liron

Tal Liron

Principal Software Engineer, Red Hat
Tal has been working in the cloud orchestration space for several years. He is currently in the networking team of CTO office of Red Hat, where he focuses on ONAP integration, and previously was at Cloudify. He is the creator of the Puccini TOSCA compiler and has worked on the ARIA... Read More →

Friday April 5, 2019 2:30pm - 3:00pm PDT

3:10pm PDT

Cloud Native End-to-End NFV Stack - Srinivasa Addepalli, Intel Corporation & Ravi Chunduru, Verizon
To realize the full promise of NFV, some service providers have started to embrace principles of cloud-native architectures. Cloud-native architectures include realization of network functions as virtual machines, containers and Serverless functions. Kubernetes is one of the key enabling technology for deploying cloud-native applications. In this presentation, we will explore the needs of cloud-native NFV, the challenges of Kubernetes and CNI technologies in realizing NFV, works that is happening in various open source communities (LFN, CNCF) to address the challenges, and finally present the end-to-end cloud native NFV stack. At the end, we will demonstrate how various workload types such as VMs and containers can co-exist and communicate with each other by showcasing an use case with ONAP, Kubernetes, Multus, OVN, Docker and Virtlet.

avatar for Srinivasa Addepalli

Srinivasa Addepalli

Sr. Principal Engineer, Intel Corporation
Srini Addepalli is a Sr. Principal Engineer in NEX/NPG business unit of Intel Corporation. He is one of the principal architects of networking, security & Edge technologies for the Network Function Virtualization/Containerization (NFV/NFC) and Software Defined Networks (SDN). Srini... Read More →
avatar for Ravi Chunduru

Ravi Chunduru

Associate Fellow, Systems Engineering and Tech Strategy, Verizon
Ravi Chunduru is a Senior Architect at Verizon responsible for Product strategy and thought leadership in the domain of Virtual Network Services and MEC solutions. Ravi has been a key player in conceptualizing and delivering various products at Verizon such as VNS Application Edge... Read More →

Friday April 5, 2019 3:10pm - 3:40pm PDT

4:10pm PDT

Empowering High Performance User Space Network Stack on Arm - Trevor Tao, Arm
The user space TCP/IP stack based on DPDK surpasses the kernel slow path and provides much higher networking performance than the legacy kernel network stack.

In this presentation we would like to give out our work of the user space network stack on Arm, which covers the following but not limited to:
1. Introduction to high performance network stacks available on arm, which includes F-stack. mTCP, VPP Hoststack, Seastar, Rumpkernel, etc.
2. Enablement and improvement of F-Stack on arm, which is a complete Freebsd based full user space network stack. New feature development, such as IPerf3 development for F-stack and its application development ways are also given.
3. Performance evaluation and analysis to F-Stack;
4. Detailed comparisons between stacks on arm;
5. Mechanism for developing an independant network library to F-Stack;
6. Future work on use space TCP/IP stack for arm

avatar for Trevor Tao

Trevor Tao

Staff Software Engineer, Arm
Trevor Tao(Zijin Tao) is a Ph.D in Computer Networking, who has worked in this area for more than 15 years. He has worked as a network engineer in research institute of university for more than 10 years. Then he worked in IBM for almost 5 years for SDN and Cloud Networking. Now he... Read More →

Friday April 5, 2019 4:10pm - 4:40pm PDT