Networking Group

    Robotron: Top-down Network Management at Facebook Scale
    by Yu-Wei Eric Sung, Xiaozheng Tie, Starsky H.Y. Wong, James Hongyi Zengمنذ حوالي ‏3‏ أسابيع
    Networking Group
    Blog
    We developed a state-of-the art system named Robotron to manage tens of thousands of network devices connecting hundreds of thousands of servers globally at Facebook. This week, we are presenting an...
    IPv6: It's time to get on board
    by Paul Saabمنذ عام تقريبا
    Networking Group
    Blog
    From better performance and engagement to a larger addressing space, IPv6 has a host of technical benefits that should cause people to take a second look.
    Facebook Open Switching System ("FBOSS") and Wedge in the open
    by Adam Simpkinsمنذ حوالي ‏2‏ سنتين
    Networking Group
    Blog
    Today we are releasing our Facebook Open Switching System (FBOSS) project on GitHub and contributing the specification for the Wedge top-of-rack switch to the OCP networking project.
    Introducing Proxygen, Facebook's C++ HTTP framework
    by Daniel Sommermann, Alan Frindellمنذ حوالي ‏2‏ سنتين
    Networking Group
    Blog
    We are excited to announce the release of Proxygen, a collection of C++ HTTP libraries, including an easy-to-use HTTP server. In addition to HTTP/1.1, Proxygen (rhymes with “oxygen”) supports SPDY/3...

    Highlights

    Robotron: Top-down Network Management at Facebook Scale
    by Yu-Wei Eric Sung, Xiaozheng Tie, Starsky H.Y. Wong, James Hongyi ZengAugust 24, 2016
    Blog post
    Robotron: Top-down Network Management at Facebook Scale
    by Yu-Wei Eric Sung, Xiaozheng Tie, Starsky H.Y. Wong, James Hongyi ZengAugust 23, 2016
    Publication
    NetNORAD: Troubleshooting networks via end-to-end probing
    by Aijay Adams, Petr Lapukhov, James Hongyi ZengFebruary 18, 2016
    Blog post
    Milestones and plans for the OCP Networking Project
    by Omar BaldonadoNovember 3, 2015
    Blog post

    About Networking Group

    Facebook was built to help nearly 1.5 billion people connect and share, and over the last decade our tools have played a critical part in changing how people around the world communicate with one another. At this scale and rate of growth, we face new challenges on a daily basis. As Facebook grows and evolves, new network workloads arise with increasingly demanding performance and reliability requirements. Our mission is to make it possible to support such workloads. At Facebook, we are constantly building the next generation systems that will be scalable, efficient, reliable and easy to use.

    From Layer 1-4, we write the software that helps build, deploy, control, and monitor the entire Facebook network, connecting all hundreds of thousands of servers in our fleet to each other and to the Internet. Traditional networking doesn’t work well at this scale, so we write a lot of software to make the network work. We work closely with Facebook’s datacenter, backbone, corporate, and operations teams. We also work with the Traffic team on the network within the POPs. Increasingly, we provide network services to other infrastructure and engineering groups across Facebook.

    From Layer 4-7, we build global and local load balancers, acceleration proxies, RPC frameworks, content delivery networks, and mobile networking stacks, plus the automation machinery that’s required to operate an internet scale edge network and handle all of Facebook’s DNS and HTTP traffic. We operate a large network of POPs around the world that we use to accelerate site traffic and cache CDN content. We glue these varied network services together in order to orchestrate the delivery of bits from our servers to your phone or desktop.

    Aside from internal impact, we contribute to the academic community as well. We publish our work in top-tier scientific conferences, such as SIGCOMM and NSDI. Following the Facebook culture of openness and sharing, we also contribute several of its projects to the open source community. Projects like Thrift, Proxygen, FBOSS and OpenBMC were developed by us and are now used by different organizations.

    Publications

    Robotron: Top-down Network Management at Facebook Scale
    Yu-Wei Eric Sung, Xiaozheng Tie, Starsky H.Y. Wong, James Hongyi Zeng
    SIGCOMM
    Aug 23, 2016
    In this paper, we present Robotron, a system for managing a massive production network in a top-down fashion.
    Inside the Social Network’s (Datacenter) Network
    Arjun Roy, James Hongyi Zeng, Jasmeet Bagga, George Porter, Alex C. Snoeren
    SIGCOMM ’15
    Aug 17, 2015
    Large cloud service providers have invested in increasingly larger datacenters to house the computing infrastructure required to support their services.
    Wormhole: Reliable Pub-Sub to Support Geo-replicated Internet Services
    Yogeshwer Sharma, Philippe Ajoux, Petchean Ang, David Callies, Abhishek Choudhary, Laurent Demailly, Thomas Fersch, Liat Atsmon, Andrzej Kotulski, Sachin Kulkarni, Sanjeev Kumar, Hu Li, Jun Li, Evgeniy Makeev, Kowshik Prakasam, Robbert van Renesse, Sabyasachi Roy, Pratyush Seth, Yee Jiun Song, Kaushik Veeraraghavan, Benjamin Wester, Peter Xie
    12th USENIX Symposium on Networked Systems Design and Implementation (NSDI ’15)
    May 6, 2015
    Wormhole is a publish-subscribe (pub-sub) system developed for use within Facebook’s geographically replicated datacenters. It is used to reliably...

    Blog

    Robotron: Top-down Network Management at Facebook Scale
    by Yu-Wei Eric Sung, Xiaozheng Tie, Starsky H.Y. Wong, James Hongyi ZengAug 24, 2016
    NetNORAD: Troubleshooting networks via end-to-end probing
    by Aijay Adams, Petr Lapukhov, James Hongyi ZengFeb 18, 2016
    IPv6: It's time to get on board
    by Paul SaabSep 14, 2015

    Join Us

    Do you want to help more than a billion people all over the world connect and share?

    View Open Positions

    Code

    Learn about our open source tools and technologies, our challenging scaling experiences, and more.

    Go to Facebook Code