The Linux Foundation, Dell-EMC and Cisco are co-hosting a one day hackathon event. The overall theme of Innovation via Open Code and Integrations was selected to achieve the goals of wide community participation, jump start projects/efforts and integrate open networking elements within interesting solutions.
Event Details
The hackathon will be held Monday, April 3 from 8:00 am - 6:00 pm at the Santa Clara Convention Center in Grand Ballroom C. Registration for the hackathon is required, but there is no cost associated. Space is available on a first come, first serve basis. Don't miss your opportunity to participate - add this hackathon to your registration today!
5 Primary Zones
Experts and thought leaders will be on hand during the event for these zones. Teams and projects are not limited to the zones and projects. If you have any interesting ideas involving Open Source Networking, we welcome your participation!
Open Source Projects of Interest: ODL, ONOS, ONAP, OPX, OPNFV, OpenContrail, Nuage, and Open-O
ParticipationAt the conclusion of the hackathon, teams will have the opportunity to present their work. Prizes will be given - but this is not a competitive event!
Communications service providers (CSPs) have a wide range of options when building virtualized services from the ground up including multiple choices for each functional block in the ETSI NFV reference architecture. CSPs prefer heterogeneous systems with building blocks from different vendors including open source software; for such deployments interoperability becomes a crucial requirement.
OpenStack, as the NFVI and VIM, serves as a widely used cloud platform for telecom and NFV use cases. As a common base, OpenStack offers the means for vendors and other open source projects to ease the interoperability challenge by providing a set of open API’s while focusing on upgradeability and backward compatibility.
However, when it comes to productization, interoperability testing often falls short and is sometimes left to the carrier as shown by the testing programs actively run by no fewer than 10 organizations today.
Join Carsten Rossenhövel from the European Advanced Networking Test Center (EANTC) and the rapporteur (editor) of ETSI’s NFV interoperability standards, Ildikó Váncsa from the OpenStack Foundation, and Chris Price, Ericsson and OpenStack board director to learn more about
The ETSI NFV Release 2 interoperability testing activities - standardization and recently completed ETSI PlugTest. Over 40 commercial and open source implementations were tested for interoperability, including 20 virtual network functions, 10 management and orchestration solutions and 10 NFV platforms.
The New IP Agency (NIA) interoperability testing campaigns of commercial NFV implementations executed by EANTC, focusing on results, lessons learned and recommendations.
How vendors and open source projects are stepping up to the challenge, realizing they must work together.
How to stay up-to-date with OpenStack releases and the community.
How to get involved to ensure you are aware of the latest developments and contribute what you need to OpenStack.
What will I learn from attending this session?
CSPs, open source projects and vendors alike will learn more about the recent ETSI PlugTest and NIA-commissioned interoperability testing, their results and how to architect full NFV solutions that will work together. Interoperability API tests and associated marks from OpenStack will be covered, as well as features to help stay current on OpenStack releases. Attendees will also hear from Ericsson about a vendor’s point of view, and how other projects such a OPNFV are evolving and expanding in scope to address this challenge.
FD.io VPP Architecture and demo : Ray Kinsella, Network Software Solutions Architect , Intel (45 minutes)
9:00-9:15
Keynote: Build All-Cloud Networks - Bill Ren, Vice President of Network Solutions Industry & Ecosystem Development, Huawei
9:15-9:45
Architecture: Fully Automated Peer Service Orchestration of Cloud and Network Resources Using ACTN - Young Lee, Technical Director, Network Architecture of SDN, USA IP Research & Standards, Huawei
9:45-10:15
Architecture: Realizing ICN as a Network Slice for Mobile Data Distribution - Ravi Ravindran, Principal Staff Researcher, America Network Technology, Huawei
10:15-10:45
Container Networking
Design high effective network for AppChain - Wei Xu, Senior Architect, Cloud Network Lab, Huawei
10:45-11:15
Network Intelligence: ThinkNET-Experiential Networked Intelligence - Shucheng Liu, Senior Researcher & Standard Representative, Project Manager, Network Research Dept, Huawei
11:15-11:45
Data Plane: NPC++:a network programming framework for hardware acceleration - Jun Liu, Chief Architect, IP Forwarding, Huawei
Tzvika Naveh, from Amdocs will discuss the service design and onboarding process within ONAP, with a focus on the Service Design and Create (SDC) module. He will also take the audience through a short demo, showing how new enterprise services can be designed, on-boarded and made available through the ONAP platform.
TIA and its NFV Security Working Group are proud to be partnering with Linux foundation to discuss securing the transformation of tomorrow's telecomm networks. By building network functions within a strategically developed network infrastructure framework, services can meet speed-of-market demands and streamline operations. Telecommunication leaders are partnering with TIA to discuss these security principals to design, build, and operate open and secure networks.
Presentations will include:
While service providers are quickly ramping up their capabilities to orchestrate Virtualized Network Functions and SDN components, many are starting to realize that work gets them only part way to their goal of growing their revenue. For that they need to offer new, lucrative services built on top of the foundation of SDN and NFV. This tutorial addresses the orchestration of services for business agility.
To this end, MEF and TM-Forum are collaborating with major service providers in a transformational initiative to standardize Lifecycle Services Orchestration (LSO) application programming interfaces (APIs) for orchestrating connectivity services across multiple networks worldwide. AT&T, Orange, Colt Technology Services, Comcast, Level 3, PCCW Global, Sparkle, Verizon, and CableLabs’ Kyrio are leading the effort. Standardized APIs are a critical step toward enabling agile, assured, and orchestrated Third Network services over more automated, virtualized, and interconnected networks globally.
This tutorial will provide attendees with practical information on network architecture, Virtualized Network Functions deployment, and the Lifecycle Service Orchestration for an end-to-end connectivity and cloud services. We will also discuss some use cases for Virtual Network Functions applications in the network, the orchestration of the services, as well as, a lesson learned during the implementation of a major multi-operator connectivity service using Carrier Ethernet, L3VPN, and SD-WAN and the complex Orchestration of such service.
Agenda:
* Welcome
* Considerations for Deploying Virtual Network Functions and Services
* Multi-domains and Multi-operators Lifecycle Service Orchestration for Rapidly Introducing New Network Functionality
* SD-WAN Tutorial: Solution Components, Architecture, and Use Cases for SD-WAN Managed Services
* General Q&A
* Closing Remarks
About the MEF:
MEF is the driving force enabling agile, assured, and orchestrated Third Network services for the digital economy and the hyper-connected world. The Third Network concept combines the agility and ubiquity of the Internet with the performance and security of CE 2.0 networks. Third Network services provide an on-demand experience with user-directed control over service capabilities and cloud connectivity. These new services are delivered over automated, virtualized, and interconnected networks powered by LSO, SDN, NFV, and CE 2.0.
MEF leverages its global 210+ network operator and technology vendor community, builds upon the robust $80 billion Carrier Ethernet services and technology market, and provides a
practical evolution to the Third Network with LSO, SDN, and NFV implementations that build upon a CE 2.0 foundation.
This session will be moderated by Ben Pfaff:
iTAP: In-network Traffic Analysis Prevention using Software-Defined Networks
Roland Meier, David Gugelmann, Laurent Vanbever (ETH Zürich)
Smoking Out the Heavy-Hitter Flows with HashPipe
Vibhaalakshmi Sivaraman (Princeton University); Srinivas Narayana (MIT CSAIL); Ori Rottenstreich (Princeton University); S. Muthukrishnan (Rutgers University); Jennifer Rexford (Princeton University)
Dapper: Data Plane Performance Diagnosis of TCP
Mojgan Ghasemi (Princeton University); Theophilus Benson (Duke University); Jennifer Rexford (Princeton University)
The Case for Making Tight Control Plane Latency Guarantees in SDN Switches
Huan Chen (UESTC and Duke University); Theophilus Benson (Duke University)
-------------------------------------------------
Traditional carriers' transport networks consist of vertically-integrated devices with vendor-proprietary interfaces, that causes "vendor lock in" environment and interferes with adopting software based control and configuration for carriers' transport networks. NTT Communications are trying to adopt disaggregation approach for them to transform our operations by integrating commoditized multi-vendor components and SDN technology.
In this presentation, we will talk about our expectations for disaggregated transport networks and its controller architecture with multiple SDN controllers including open source software. Furthermore, we will show our internal evaluation result of disaggregated transport network feasibility and discuss future development plans.
As the standards on NFV orchestration evolves slowly, and also there are many MANO open source implementation today like OSM, OPEN-O, Open-Baton, Tacker, even ECOMP. there is an increasingly considerable divergence among open source implementations, which may lead to industry fragmentation since the alignment of the Open Source and Standard on information/data modeling, which is critical to enable interoperability among MANO stack components. The invited panelists, are expected to discuss the current practice and status of collaboration between related SDOs and open source communities on NFV orchestration, which is particularly focused on identifying and addressing potential gaps between SDO specifications to open source implementations (e.g. Among ETSI NFV IFA, YANG, Tosca and OpenStack).
Containers are here. They complicate networking as there is a requirement to coexist in a hybrid environment with hypervisors and have to operate at a fairly high scale.
This panel will discuss focus on Container Networking landscape including CNI and other options available, Challenges for scaling CN beyond simple IP connectivity into L4-7 services. Discussion will also highlight the Operational aspects of containers and VMs with visibility, config management and API into virtual and physical world. Finally the future of Container Networking holistically as plug and play for both Packaging and Isolation in Greenfield and Brownfield environment will be covered.
Moderated by Xin Jin:
P4FPGA: A Rapid Prototyping Framework for P4
Han Wang (Cornell University); Robert Soulé, Huynh Tu Dang (Università della Svizzera italiana); Ki Suh Lee, Vishal Shrivastav, Nate Foster, Hakim Weatherspoon (Cornell University)
Whippersnapper: A P4 Language Benchmark Suite
Huynh Tu Dang (Università della Svizzera italiana); Han Wang (Cornell University); Theo Jepsen (Università della Svizzera italiana); Gordon Brebner (Xilinx Inc.); Changhoon Kim (Barefoot Networks); Jennifer Rexford (Princeton University); Robert Soulé (Università della Svizzera italiana); Hakim Weatherspoon (Cornell University)
Swing State: Consistent Updates for Stateful and Programmable Data Planes
Shouxi Luo, Hongfang Yu (University of Electronic Science and Technology of China); Laurent Vanbever (ETH Zürich)
P5: Constructing Efficient Programmable Switch Pipeline using Policy Intents
Anubhavnidhi Abhashkumar (University of Wisconsin Madison); Jeongkeun Lee (Barefoot Networks); Jean Tourrilhes, Sujata Banerjee, Joon-Myung Kang, Wenfei Wu (Hewlett Packard Labs); Aditya Akella (University of Wisconsin Madison)
-----------------------------------------------------------------------------
To RSVP to this session, please log in to your Open Networking Summit registration here.
If you are not registered to Open Networking Summit (ONS), you can do so here.
If you have registered to SOSR already, no action is needed.
5G is the natural evolution of today’s mobile networks, but it is far more than a faster radio access technology. Like the previous generations of mobile network standards (2G, 3G, 4G …), 5G is driven by the promise of faster and more efficient radio access technology that promises ubiquitous broadband (50Mbps everywhere), lower latency, seamless communication for mobility at higher speeds, and extreme scale. But the 5G architecture is also introducing a re-architecture of the core, incorporating new technologies and approaches that have emerged in the telco networking industry – software-defined networking (SDN), network functions virtualization (NFV), and mobile edge computing (MEC) to name a few.
There are several industry-wide open development efforts relevant to many of the component technologies and architectures mentioned above. But given the focus on performance, scale, reliability and manageability for 5G networks, can open development and open networking be relevant here? The challenge is to design a common network infrastructure that can serve disparate use cases. The diversity of requirements necessitates a network architecture that is flexible, programmable, intelligent and distributed. This panel will discuss the question of how we can harness the power of openness to achieve this goal. The panel will cover the point of view both of owners and operators of 5G networks, as well as solution, software and system providers.
Sample questions (exemplary only):
· What constitutes a 5 G network for you – do you consider it as a better radio access, or something more?
· Are you in the process of transforming your network? How is that related to 5G?
· Do you see a relationship between the network transformation being undertaken by CSPs and 5G networks, and what is it?
· What business problems are you attempting to solve by transitioning to 5G networks?
· Do you see 5G enabling new devices or new customer sets for you?
· A lot of emphasis has been placed on standards and standardization for 5G networks. Is there a role for open development and open source here?
· How do you balance the stringent requirements that 5G networks are expected to address, and the nature of open development?
Building scalable cloud applications as a collection of micro-services is a promising approach to NFV Orchestration. This is because micro-services are well established as the cornerstone of DevOps, which operators see as a path to improving agility. But organizing a collection of VNFs as micro-services is not sufficient. There are two issues. The first is that micro-services assume a single domain of trust, but network operators must accommodate multiple actors. The second is that micro-services isolate information, but orchestration requires global visibility across a set of micro-services. This talk describes a service control plane that is layered on top of a collection of micro-services to address these two limitations. In addition to describing the role of a service control plane, this talk reports our experiences implementing and using a service control plane in CORD.
Dis-agregration is real… This trend started with SDN and the separation of Data plane and Control plane. The scope has expanded to include separate of hardware and software and created a whole new industry of white boxes, general purpose X86 commodity hardware. All three markets - Cloud, Enterprise and Carriers are now engaged in various solutions inside the Data Center. The disaggregation is impacted all parts of the network including Access and Edge layers.
Panel covers latest advances in OS, Hardware and Software to enable a full interoperable ecosystem.
This session will be moderated by James Zeng:
Piggybacking Network Functions on SDN Reactive Routing: A Feasibility Study
Chang Liu, Arun Raghuramu, Chen-Nee Chuah (University of California, Davis); Balachander Krishnamurthy (AT&T Labs-Research)
ParaBox: Exploiting Parallelism for Virtual Network Functions in Service Chaining
Yang Zhang (University of Minnesota, Twin Cities); Bo Han, Bilal Anwer, Vijay Gopalakrishnan, Joshua Reich, Aman Shaikh (AT&T Labs -- Research); Zhi-Li Zhang (University of Minnesota, Twin Cities)
--------------------------------------------------------------------------------SOSR Posters Program
MPVisor: A Modular Programmable Data Plane Hypervisor
Cheng Zhang, Jun Bi, Yu Zhou, Abdul Basit Dogar, Jianping Wu (Tsinghua University)
PRAM: Priority-aware Flow Migration Scheme in NFV Network
Zili Meng, Jun Bi, Chen Sun, Anmin Xu (Tsinghua University); Hongxin Hu (Clemson University)
Gotthard: Network Support for Transaction Processing
Theo Jepsen, Leandro Pacheco de Sousa, Huynh Tu Dang, Fernando Pedone, Robert Soulé (Università della Svizzera italiana)
Meticulous Measurement of Control Packets in SDN
Yash Sinha (Birla Institute of Technology and Science Pilani); Shikhar Vashishth (Indian Institute of Science Bangalore); K. Haribabu (Birla Institute of Technology and Science Pilani)
Enabling Fast Hierarchical Heavy Hitter Detection using Programmable Data Planes
Diana Andreea Popescu, Gianni Antichi, Andrew W. Moore (University of Cambridge)
Towards an OS for the Network Data Plane
Wei Zhang (George Washington University); Abhigyan Sharma, Kaustubh Joshi (AT&T Labs Research); Timothy Wood (George Washington University)
PVPP: A Programmable Vector Packet Processor
Sean Choi (Stanford University); Xiang Long (Cornell University); Muhammad Shahbaz (Princeton University); Skip Booth, Andy Keep, John Marshall (Cisco); Changhoon Kim (Barefoot Networks)
Delorean: Using Time Travel to Avoid Bugs and Failures in SDN Applications
Zhenyu Zhou, Theophilus Benson (Duke University); Marco Canini (KAUST); Balakrishnan Chandrasekaran (TU Berlin)
Siren: A Platform for Deployment of VNFs in Distributed Infrastructures
Lyndon Fawcett, Nicholas Race (Lancaster University)
Automating SDN Composition: A Database Perspective
Anduo Wang (Temple University); Jason Croft (University of Illinois at Urbana-Champaign)
Magneto: Unified Fine-grained Path Control in Legacy and OpenFlow Hybrid Networks
Cheng Jin (University of Minnesota); Cristian Lumezanu, Qiang Xu (NEC Laboratories America); Hesham Mekky, Zhi-Li Zhang (University of Minnesota); Guofei Jiang (NEC Laboratories America)
SDX-Based Flexibility or Internet Correctness? Pick Two!
Rüdiger Birkner (ETH Zürich); Arpit Gupta, Nick Feamster (Princeton University); Laurent Vanbever (ETH Zürich)
ParaBox: Exploiting Parallelism for Virtual Network Functions in Service Chaining
Yang Zhang (University of Minnesota, Twin Cities); Bo Han, Bilal Anwer, Vijay Gopalakrishnan, Joshua Reich, Aman Shaikh (AT&T Labs Research); Zhi-Li Zhang (University of Minnesota, Twin Cities)
SOSR Demos Program
Demo: Adversarial Network Forensics in Software Defined Networking
Stefan Achleitner, Thomas La Porta, Trent Jaeger, Patrick McDaniel (Pennsylvania State University)
Demo: SDN-enabled Traffic Engineering and Advanced Blackholing at IXPs
Christoph Dietzel (TU Berlin / DE-CIX); Gianni Antichi (University of Cambridge); Ignacio Castro, Eder L Fernandes (Queen Mary, University London); Marco Chiesa (Université catholique de Louvain); Daniel Kopp (DE-CIX)
Demo: IPv6 Segment Routing to the End Host: A Linux Kernel Implementation
David Lebrun (Université catholique de Louvain)
Demo: Implementing iptables using a programmable stateful data plane abstraction
Luca Petrucci, Marco Bonola, Salvatore Pontarelli, Giuseppe Bianchi (CNIT / University of Rome Tor Vergata); Roberto Bifulco (NEC Laboratories Europe)
Demo: Programming Distributed Control Planes (invited)
Ryan Beckett (Princeton University); Ratul Mahajan, Jitendra Padhye (Microsoft); Todd Millstein (UCLA); David Walker (Princeton University)
Please join us on Monday, April 3 at 5:30pm at the Hyatt Regency Santa Clara, Napa Rooms, for a ONF meetup and reception. This event is open to ONF Members and non-members - All are welcome!
ONF Executive Director Guru Parulkar and members of the ONF Board of Directors will be in attendance for a short presentation and Q&A followed by drinks and Hors d'oeuvres.
Please RSVP here.
Standards have long played an important role in the telecommunications industry, but there is increasing recognition that open source software is the key to rapid innovation. This realization is the impetus behind the recently announced plans for ONF and ON.Lab to merge, where the goal is to “simultaneously prototype-and-prove new solutions while laying the groundwork for those solutions to interoperate with each other.” It is this combination of agility and interoperability that the new ONF hopes will provide the best opportunity to grow and accelerate the adoption of SDN.
This is reminiscent of the IETF’s original "rough consensus and running code" mantra, but thanks to SDN, much has changed in how we build networks. This talk outlines our plan to combine the best of ONF and ON.Lab, and our experiences to-date pursuing Software Defined Standards.
Intent-based networking has gained a lot of interest in recent years with several different open source communities (including ONF, OpenDaylight and ONOS). However, network diagnosis and troubleshooting remain two essential aspects of network management that still require massive manual effort and extensive expert knowledge. So far, no approach has focused on troubleshooting the network at the intent level. We argue that providing reasoning capabilities about why an observed network state happens according to specified input policies can help simplify this complexity. In this talk, we present our work on network troubleshooting at the intent level. Our solution provides capabilities to enable posing what-if type questions at the policy level –e.g. when a new input policy is introduced– to help answer reachability questions both at the policy and infrastructure level.
In today’s world we are moving towards a world where things are connected and automated. We are trying to ease our daily living with the help of technology through Internet. One such thing is Assisted Living where technology provides long term senior care in terms of Health, Food, Transportation, Hospitality, etc.,. It also helps physically challenged persons. The session provides how to build a system for Assisted Living with the architecture of Internet of Things (IoT). The speaker will demonstrate few IoT hardware platforms, sensors technologies related to Assisted Living such as Moisture, Gas, and Buttons, Temperature sensors, Seeeduino Arch Pro and IoT cloud platform. At the end presenter will share few business cases of Assisted Living projects for elderly care including: Smart Locking System, Vibration Alert for Hearing Impaired, Smart Helmet, Smart Help, Vision Band.
One of the major challenges and requirements in achieving a very high (>99.99%) reliability of operation of any major network infrastructure (i.e. data center, enterprise, campus, etc.) is the ability to design and deploy an always-on active system that performs end-to-end functional testing of all the network-connected infrastructure components and, as a result, monitors the infrastructure and its dependent external services with high accuracy and granularity (down to the packet level) in the most efficient way; consuming the least amount of computational or network resources.
When it comes to packet loss detection, metrics reported by the original manufacturers cannot be relied upon; their tools may either be buggy or, in most cases, do not provide APIs for extracting measurements. Therefore, we needed to create our own tool; this is the gap Arachne is filling.
In this talk, we present Arachne. Arachne is a packet loss detection system and an underperforming path detection system. It provides fast and easy active end-to-end functional testing of all the components in Data Center (DC) and Cloud infrastructures. Arachne is able to detect intra-DC, inter-DC, DC-to-Cloud, and DC-to-External-Services issues by generating minimal traffic.
Data plane has been a key component of Networking for Decades. It is going through massive disruption and transition as major trends like Disaggregation, Programmable hardware and SDN take center stage.
The intersecting technology requirements of 5G, Webscale data centers and SDN/NFV has shown new innovations in the area of Silicon, hardware, data plane acceleration using general purpose servers.
This panel will bring out ideas and actual innovation that is in the pipeline to create the fastest, most open, programmable and scalable data plane from silicon to hardware to low level software.
The NFV movement has embraced “openness” as its central tenet, spawning multiple open source orchestration efforts. However, the focus on openness has left several important goals unaddressed, such as operational simplicity, state-of-the-art performance, and advanced development tools for NFs. In this talk, Scott Shenker and Sylvia Ratnasamy will survey the state of NFV and point to a new approach which retains the important goal of openness, which is preventing vendor lock-in, while providing much better simplicity, performance, and development tools than the current NFV ecosystem has yet achieved.
Colt is transforming the way they do business and offer network services to customers through the adoption of SDN & NFV as part of a company-wide transformation program called Novitas. Javier Benitez will focus in this presentation on active Colt’s Novitas developments, sharing with the audience their experience in deploying SDN & NFV solutions in production both for Ethernet and IP services, the learning associated as well as their future plans. In particular, Javier Benitez will cover Colt developments around Ethernet & IP on Demand, SD VPN, SDN controlled MPLS packet core and SDN/NFV NNI standardization.
Most, if not all, modern network controllers acknowledge that they are distributed systems and must deal with how to replicate state. There is a trade-off between strong consistency (with easier debugging, programming models, and understandability) on the one hand and tolerance for high latency between controllers and possible partitions on the other hand. To address this, we have extended a strongly-consistent data store (based on RAFT) to also allow for best-effort replication to additional nodes. This allows there to be a small number of nodes (with low latency between them) providing a strongly-consistent data store, while efficiently replicating state to other entities including other controllers acting as warm standbys to handle disaster recovery. Future extensions can allow simple federation by leveraging selective best-effort replication to exchange state without tight coupling.
This session will be moderated by Boon Thau Loo:
Concise Encoding of Flow Attributes in SDN Switches
Robert MacDavid (Princeton University); Rüdiger Birkner (ETH Zürich); Ori Rottenstreich, Arpit Gupta, Nick Feamster, Jennifer Rexford (Princeton University)
Magneto: Unified Fine-grained Path Control in Legacy and OpenFlow Hybrid Networks
Cheng Jin (University of Minnesota); Cristian Lumezanu, Qiang Xu (NEC Laboratories America); Hesham Mekky, Zhi-Li Zhang (University of Minnesota); Guofei Jiang (NEC Laboratories America)------------------------------------------------------------------------------
To RSVP to this session, please log in to your Open Networking Summit registration here.
If you are not registered to Open Networking Summit (ONS), you can do so here.
If you have registered to SOSR already, no action is needed.
The path towards open networking among service providers and enterprise is not as smooth, primarily due to the existence of huge vendor-specific software and hardware. In this presentation, Ali Tizghadam will explain how TELUS has overcome the issue by introducing a customized open data dissemination platform. The platform enables service providers to write customized plugins for their legacy data collectors (brown field) and at the same time enables them to embrace data collectors available in open community for green field development. The platform publishes data to a common broker network, where one can subscribe to specific data and receive new updates. Moreover, the platform offers APIs for on demand access to the data in standard way. Using this approach, TELUS has abstracted the legacy layer from control layer and all controllers / orchestrators use one unified data access method.
As the mobile traffic carried by cellular networks has been growing rapidly and the networks gets bigger and more complex, network operators have been forced to search for solutions to substantially enhance network visibility. This talk introduces SKT integrated Network Analyzer (TiNA) and Converged Appliance Platform (T-CAP) which help us improving the efficiency of network operation, troubleshooting, and analyzing traffic. TiNA is composed of virtual network packet broker, flow analyzer, high speed packet dump system, connection performance analyzer, and 3D-based network management system. T-CAP is an open architecture of a server-switch type hardware. We will review how to implement those TiNA functions based on open source (e.g., DPDK, Spark Streaming) and T-CAP. Finally, we will also discuss about the use-cases of TiNA and T-CAP for the private cloud & telco network infrastructure.
SDN is at the foundation of all large scale networks in the public cloud, such as Microsoft Azure - at past ONSes, Microsoft has detailed how all of Azure's virtual networks, load balancing, and security operate on SDN. But how do we make a software network scale to an era of 40, 50, and 100 gigabit networks on servers, providing great performance to end customers with ever increasing VM and container scale and density?
In this presentation, Daniel Firestone and Gabriel Silva will detail Azure Accelerated Networking, using Azure's FPGA-based SmartNICs. They will show how using FPGAs, we can achieve the programmability of a software network with the performance of a hardware one. They will detail how this and other host SDN advances have led to huge performance increases for Linux VMs in particular, and Linux-based NFV appliances, giving Azure industry-leading network performance.
How does a feature idea or user request become a working piece of an open source project? Can a developer just write some code and release it to the world?
If you're a user, or a developer new to the open source world, this talk will give you a clear understanding of the structured process by which an idea becomes usable code in a well-run open source project, from scope review to code acceptance. During this talk, Phil Robb will also shed light on the typical timelines, success factors for both upstream features and downstream commercial offerings, and best practices in managing cross-organizational dynamics for the good of all.
This session will be moderated by Colin Dixon, Brocade:
Decentralized Consistent Updates in SDN
Thanh Dang Nguyen (University of Chicago); Marco Chiesa (Université catholique de Louvain); Marco Canini (KAUST)
Adversarial Network Forensics in Software Defined Networking
Stefan Achleitner, Thomas La Porta, Trent Jaeger, Patrick McDaniel (Pennsylvania State University)
NEAt: Network Error Auto-Correct
Wenxuan Zhou, Jason Croft, Bingzhe Liu, Matthew Caesar (University of Illinois at Urbana-Champaign)
SDX-Based Flexibility or Internet Correctness? Pick Two!
Rüdiger Birkner (ETH Zürich); Arpit Gupta, Nick Feamster (Princeton University); Laurent Vanbever (ETH Zürich)
"Let me rephrase that": Transparent optimization in SDNs
Santhosh Prabhu, Mo Dong, Tong Meng, Brighten Godfrey, Matthew Caesar (UIUC)
BigBug: Practical Concurrency Analysis for SDN
Roman May, Ahmed El-Hassany, Laurent Vanbever, Martin Vechev (ETH Zürich)
-----------------------------------------------------------------------
To RSVP to this session, please log in to your Open Networking Summit registration here.
If you are not registered to Open Networking Summit (ONS), you can do so here.
If you have registered to SOSR already, no action is needed.
While innovative, leading edge open source efforts are happening more and more often. Cleaning up after such forward progress is often forgotten. Consolidation and interoperability of the SDN controller and data plane domains remain an elusive target. There is way too much non-differentiated value in the ecosystem and not enough developers to warrant the diversity. As the open source initiatives associated with SDN controller and data plane become increasingly abundant we would like to take this opportunity to discuss the benefits of both diversity and consolidation in this domain. The recent emergence of the ODL controller with an FD.io forwarding plane to address carrier needs provides a new compelling option.
Open source adoption is quite mature at the application level, often to the point of invisibility. The use of open source in infrastructure is far newer, and subject to many myths and much confusion, from expected cost savings to technical investment required. In the absence of understanding how to evaluate and consume open source options, many enterprises have defaulted to wholly proprietary solutions from their existing vendors, in spite of their desire for greater purchasing flexibility. In this talk, Lisa Caywood, OpenDaylight Director of Ecosystem Development, will discuss:
- How to evaluate the health and outlook of an open-source infrastructure project
- What "free" means?
- Acquisition options, from download/compile to shrink-wrapped solutions
- How and why to work with vendors
- What to look for in commercial open source solutions
- Advantages of upstream involvement
There is a widespread impression in the network engineering world that centralized control planes are simpler than distributed control planes. This impression often drives the perception of the future of network engineering and design in a way that favors centralized controller based systems over more traditional distributed control planes. This session will investigate this claim by considering centralized and distributed control planes in the light of two models; a network complexity model, and the CAP theorem. The proposed outcome is a more realistic view of complexity in centralized systems, and a possible way forward in thinking through where each kind of control plane is useful.
This presentation makes five points:
1. Software innovations have become central to business operations. obtaining high performance software, such as Open Source Software, is nearly zero if a firm uses an online exchange such as GitHub and has access to a permissive license. This speeds up software development and reduces its cost because firms can often avoid paying licensing costs.
2. Software innovations are employed in such a wide range of industries that software should be a General-Purpose Technology.
3. Software behaves unlike other products. It is part of a continuing development process with developers adding code as needed. This ongoing process can be seen in new Internet-related software, such as Linux, Open Source, and Docker/containers.
4. The time it takes to complete software innovations is dropping sharply.
5. Businesses take swift advantage of software innovations
Join us for happy hour from 5.30 to 7 pm in the Bayshore Room at the Hyatt . Be sure to stop by 6WIND’s booth, #325, for an entry ticket.
Broadband is mandatory for the economic success of communities around the globe. Yet, the economics of last mile fiber networks are daunting. To be successful, service providers of all ilk need business and technical architectures that address real-world challenges of the access network. Open source initiatives, such as CORD project, have emerged as a technical architecture that provides an open platform to support a wide range of today’s access network business models. The business of access is changing however. Consumer demands for higher speeds, cord cutting, cities building their own broadband network, new last mile entrants such as GoogleFiber, edge processing and emerging wireless technologies are all impacting the business of access. This presentation will discuss the challenges of broadband, the current broken broadband ecosystem and how open source leads to Open Broadband Access, a technical and business architecture that benefits the entire ecosystem and creates a platform for innovation.
SD-WAN is a hot technology that is moving from the drawing board to production. It has changed the WAN equation forever and resulted in enterprises looking at the Connectively and Carrier Services quite differently.
It is an SDN use case to connect enterprise locations over large distances. It promises to reduce enterprises’ IT expenses by using broadband connections and running managed services in the cloud.
It simplifies the management and operation of a WAN by decoupling the networking hardware from its control mechanism. Panel will discuss details of use cases that standards and Operators are deploying e.g. including major multi-operator MEF reference implementation of Orchestrated L3VPN.
Are we close to hitting peak confusion regarding containers? Are we in the middle of a container war? Does anyone really know how to build a cloud native computing architecture? What does it mean to be cloud native?
The questions are many as we build a new computing paradigm for enterprises and service providers. With every answer we provide to the big challenges cloud operators and application developers face, three more are raised.
All Open Source Networking project depend on having access to a Universal Dataplane that is:
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. The recently announced fourth release, Danube, represents a growing maturity for both the project and upstream partners and brings together elements across the stack to more quickly introduce technologies that meet the needs of operators. Join Tapio Tallgren, OPNFV TSC Chair, and Heather Kirksey, OPNFV Director, to learn how Danube, Euphrates, and the many rivers to come are helping to build the next-generation network for NFV.
OpenDaylight has become a nexus for open source integration, creating a new open networking stack and enabling a new generation of open source, agile IT infrastructure. The fifth “Boron” release provides new tooling and documentation to support application developers, as well as greater integration with industry frameworks from OPNFV and OpenStack to CORD and Atrium. Boron also brings a practical focus on two leading types of deployments: (1) direct control of virtual switches to provide network virtualization and NFV and (2) management and orchestration of existing networks to provide new features and automation. This talk will cover trends in open SDN and cloud networking, with a focus on Boron milestones. In particular, it dives into the architecture across OpenStack and OpenDaylight to enable OpenStack service function chaining support in OpenDaylight.
10:30am - 11:00am
ONAP Welcome - Message from the Chairman of the Board, Chris Rice, AT&T; President of the Board, Yachen Wang, China Mobile
11:00am - 11:30am
Why ONAP: Requirements and Use Cases from End Users - Mazin Gilbert, AT&T; Vincent Danno, Orange; Yachen Wang & Lingli Deng, China Mobile
11:30am - 12:30pm
Targeted Architecture Requirements & High Level Roadmap - Phil Robb, OpenDaylight; Lingli Deng, China Mobile; Vimal Begwani, AT&T
** There will be a lunch break from 12:30pm - 1:30pm**
1:30pm - 2:15pm
Modeling, VNF Guidelines & SDKs - Steven Wright, AT&T, Lingli Deng, China Mobile, Hui Deng, Huawei
2:15pm - 2:30pm
Microservice Powered Orchestration- Huabing Zhao, ZTE
We are currently witnessing a major shift in the networking landscape which requires industry stakeholders to not only rapidly create specifications and code, but to also ensure that they avoid inadvertently forming new technological and operational silos that leave the market fragmented and unable to scale to meet the challenges of tomorrow. This session will be threefold and present how MEF innovates with its development model and how it interacts with other industry stakeholders.
Introduction to MEF and to MEF LSO - SDNFV Service Orchestration (10 min) - Dan Pitt, Senior Vice President, MEF
MEF and TM Forum are collaborating with major service providers such as AT&T, Orange, Colt Technology Services, Comcast, Level 3, PCCW Global, Sparkle, Verizon, and CableLabs’ Kyrio in a transformational initiative to standardize Lifecycle Services Orchestration (LSO) application programming interfaces (APIs) for orchestrating connectivity services across multiple networks worldwide. This session will provide an introduction to the MEF LSO and its APIs.
Connecting Open Source & Standards - A New Agile Development Model (20 min) - Pascal Menezes, CTO, MEF
MEF has adopted a more agile development model that involves realizing specifications in code more quickly, testing specifications out in reference implementations involving open and closed source solutions, and providing a feedback loop for further specification development. MEF has created OpenLSO (Open Lifecycle Service Orchestration) and OpenCS (Open Connectivity Services) reference implementations and is working with prominent open source projects and member companies in order to maximize alignment of market implementations with MEF published and emerging LSO and connectivity services specs. In addition, the MEF LSO Hackathons bring together software developers and network experts for hands-on development of OpenLSO and OpenCS implementations. This first part of the presentation will cover the different aspect of connecting Open Source and Standards.
The Orchestration Landscape: Open Source Projects & Specifications (20 min) - Rami Yaron VP, Strategy & Business Development, Telco Systems, Global Marketing Co-chair, MEF
This part of the presentation highlights how key industry SDOs and Open Source projects address major challenges in balancing the need to innovate and provide solutions quickly in their respective fields of expertise with the need to maximize interconnectivity and interchangeability of those solutions for service providers in the long term in a very complex and rapidly changing ecosystem. How can open APIs be developed that ensure commonality of underlying information models straddling different domains of expertise? How should companies deal with the use of different but rapidly evolving tools in different organizations? How to deal with releases from different Open Source projects and SDOs at different times? These are just some of the challenges being faced today which this presentation will discuss.
Open vSwitch (OVS) is a multilayer open source virtual switch. OVS is designed to enable massive network automation through programmatic extension, while still supporting standard management interfaces. OVN is a new network virtualization project that brings virtual networking to the Open vSwitch user community. OVN includes logical switches and routers, security groups, and L2/L3/L4 ACLs, implemented on top of a tunnel-based overlay network.
In this presentation, we will provide an overview of the current state of the projects and their future plans, such as:
OpenSwitch has gone through changes over the last year with HPE transitioning control to Dell EMC and Snaproute as the main contributors. With the change in control, there has been and updated architecture and a re-focusing of effort. OpenSwitch 2.0 (OPX) has now hit its initial milestone of having a fully functional Network OS and we would like to share with the community the new developments along with the roadmap.
OpenStack, as the NFVI and VIM, serves as a widely used cloud platform for building telecom and NFV systems.
The software is developed and delivered by a large open source community with a six month release cadence. The latest Ocata version was released the end of February and the next (sixteenth) release Pike is under development, while the community is begins planning for the following release, Queen, at the Forum at the OpenStack Summit in Boston in May. Throughout all cycles, the contributors work on both adding new features and stabilizing the platform itself.
During this session we will highlight the recently added features and evolution of OpenStack Networking including those in the Neutron service as well as in container networking. You will learn more about the roadmap along with how telecoms, open source projects and vendors can use OpenStack’s programmable infrastructure to advance network virtualization.
Join recent Neutron PTL Armando Migliaccio, OpenStack Foundation’s Ildikó Váncsa, and Paul Carver, AT&T for this detailed roadmap session covering Neutron and its plug-ins and sub-projects, containers for control plane and VNFs, container networking (Kuryr), plans for OpenStack in or working with OPNFV, several MANO projects, and more. AT&T, a leading-edge CSP, will share how they successfully leverage the OpenStack community processes, communications, and development tools to turn requirements into plans on the roadmap.
What will I learn from attending this session?
Hear from this exciting line-up of speakers to learn about the recent and planned features for OpenStack in the networking space. Find out how to keep up-to-date and how to contribute features important to your business to the platform.
DPDK is a set of software libraries and drivers which accelerate packet processing workloads running on general purpose CPUs. It’s used everywhere that high performance networking is required, including wireless core and access, cloud, enterprise, network appliances, storage, cable and more.
In this presentation we will provide an overview of the current status of the project and future plans including:
ETSI Roadmap
Pierre Lynch, Lead Technologist, xia
The session will summarize Release 2 of the ETSI NFV ISG documents and features, which were recently announced. Furthermore, it will outline the details of Release 3, which is ongoing work within the ISG. Lastly, the session will briefly summarize the ongoing work of the TST Working Group (Testing,Experimentation and Open Source), including the results of the first ETSI NFV Plugtest where multiple open source communities were actively represented.
Enabling 5G with NFV – Network Operator Perspectives
Don Clarke, CableLabs (Chair of the ETSI NFV Network Operator Council)
In October 2012, a group of 13 global network operators published an influential white paper describing their vision for the future of telecommunications networks based on technology which they called “Network Functions Virtualisation” or simply “NFV”. Simultaneously they launched the ETSI NFV Industry Specification Group which is the central body converging requirements for NFV and developing specifications which enable standards and open source communities to use NFV in a common framework. This latest NFV white paper co-authored by 23 network operators - including members of the original founding group, outlines what they see as the priorities for NFV to support 5G. This session will overview this paper and describe the rationale for the main proposals.
With the booming of Container technology, more and more options are available for enterprise users, like Docker and Kubernetes.
Although Containers provided obvious advantages: simple and faster deployment, portability and lightweight cost. But the networking challenges are significant.
Users need to restructure their network and support container deployment with current cloud framework, like container and VMs.
In this presentation, Dr. Tan will introduce new Open Source container networking solution (iCAN), which provides one management framework to configure container networking through Open/friendly interfaces, a variety of data plane components.
iCAN can not only simplify network deployment and management, but also implement flexible mechanism to validate network Service Level Agreement(SLA), which is important factor for enterprise to deliver successful and stable service via container native.
In addition, Dr.Tan will share leanings during solving networking challenges and industry view about next steps for Container base Cloud. Beside above, Dr.Tan will introduce huawei open source plan to inspire container deployment with new iCAN.
