Navigation
Search
|
xAI picked Ethernet over InfiniBand for its H100 Colossus training cluster
Tuesday October 29, 2024. 08:30 PM , from TheRegister
Work already underway to expand system to 200,000 Nvidia Hopper chips
Unlike most AI training clusters, xAI's Colossus with its 100,000 Nvidia Hopper GPUs doesn't use InfiniBand. Instead, the massive system, which Nvidia bills as the 'world's largest AI supercomputer,' was built using the GPU giant's Spectrum-X Ethernet fabric.…
https://go.theregister.com/feed/www.theregister.com/2024/10/29/xai_colossus_networking/
Related News |
25 sources
Current Date
Nov, Thu 21 - 14:53 CET
|