In this talk we provide an overview of Meta’s RDMA deployment based on RoCEV2 transport for supporting our production AI Training infrastructure. We will shed light on how we designed our infrastructure to both maximize raw performance and consistency that is fundamental for the workload. We will talk about the challenges we solved in Routing, Transport and Hardware layers we solved along the way to scale our infrastructure. We will also touch on opportunities that remain in this space to make further progress over the next few years.
- WATCH NOW
- VIEW EVENTS
- 2023
- JANUARY
- No Events
- FEBRUARY
- no events
- MARCH
- RTC @Scale 2023
- April
- no events
- May
- AI Infra @Scale
- June
- no events
- July
- Systems @Scale Summer 2023
- August
- Product @Scale 2023
- September
- Networking @Scale 2023
- Reliability @Scale 2023
- October
- Mobile @Scale 2023
- November
- Video @Scale 2023
- December
- Systems @Scale Winter 2023
- 2022
- January
- no events
- February
- RTC @Scale 2022
- March
- Systems @Scale Spring 2022
- April
- Product @Scale Spring 2022
- May
- Data @Scale Spring 2022
- June
- Systems @Scale Summer 2022
- Networking @Scale Summer 2022
- July
- no events
- August
- Reliability @Scale Summer 2022
- September
- AI @Scale 2022
- October
- no events
- November
- Networking @Scale Fall 2022
- Video @Scale Fall 2022
- December
- Systems @Scale Winter 2022
- 2021
- 2020
- January
- no events
- February
- no events
- March
- no events
- April
- no events
- May
- no events
- June
- no events
- July
- no events
- August
- Systems @Scale Remote Edition — Summer 2020
- September
- no events
- October
- no events
- November
- Performance @Scale NY 2020
- Keeping the Lights On @Scale
- AI @Scale 2020
- December
- no events
- 2019
- January
- no events
- February
- no events
- March
- no events
- April
- no events
- May
- no events
- June
- Performance @Scale 2019
- Systems @Scale Summer 2019
- July
- no events
- August
- no events
- September
- Networking @Scale California 2019
- Systems @Scale Fall 2019
- Video @Scale 2019
- October
- The @Scale Conference 2019
- November
- Fighting Abuse @Scale 2019
- Systems @Scale Tel Aviv Fall 2019
- Networking @Scale Boston 2019
- December
- no events
- 2018
- January
- Android @Scale 2018
- February
- no events
- March
- Performance @Scale 2018
- April
- Video @Scale 2018
- Fighting Abuse @Scale 2018
- May
- Networking @Scale 2018
- June
- no events
- July
- Systems @Scale Summer 2018
- August
- no events
- September
- The @Scale Conference 2018
- October
- Data @Scale Boston 2018
- November
- Mobile @Scale Tel Aviv 2018
- December
- no events
- 2017
- January
- no events
- February
- Machine Learning @Scale 2017
- Video @Scale 2017
- March
- no events
- April
- no events
- May
- Dev Tools @Scale 2017
- Networking @Scale 2017
- June
- Data @Scale 2017
- July
- no events
- August
- The @Scale Conference 2017
- September
- no events
- October
- Mobile @Scale Boston 2017
- November
- no events
- December
- no events
- 2016
- January
- Video @Scale 2016
- February
- Performance @Scale 2016
- March
- Mobile @Scale 2016
- April
- no events
- May
- Networking @Scale 2016
- June
- Data @Scale 2016
- July
- no events
- August
- The @Scale Conference 2016
- September
- no events
- October
- Boston Networking @Scale 2016
- November
- Spam Fighting 2016
- December
- no events
- 2015
- 2023
- DIVIDER
- EXPLORE TOPICS
- MACHINE LEARNING AND AI
- Data, Systems, and Networking
- MOBILE, VIDEO, AND WEB
- DEV TOOLS AND OPS, PRIVACY, SUSTAINABILITY, AND PERFORMANCE
- Fighting Abuse and Security
- DIVIDER
- Annual @Scale Conference
- Blog
- Community Forum
- Speaker Submissions
- About @Scale