Adi Gangidi
where else to find them
what they talk about
where they keep going back
- Networking @Scale2 editions2023–2024
on a stage with
- Jenya Leeonce
- KR Kishoreonce
- Pavan Balajionce
who talks about the same things
- Kaushik Veeraraghavan4 talks
- Omar Baldonado5 talks
- Aparna Ramani7 talks
- Erez Cohen3 talks
- Jim Julson3 talks
- Jitendra Padhye5 talks
| title | conference | date |
|---|---|---|
| Faster Than Fast: Networking and Communication Optimizations for Llama 3 | Networking @Scale 2024 | 2024-09-11 |
Meta builds two 24,000‑GPU clusters (AI zones of ~3,000 GPUs each, 1:7 oversubscription at aggregate switches) to train Llama 3 (8B, 70B, 405B) and reports… | ||
| GenAI Training In Production: Software, Hardware & Network Considerations | Systems @Scale 2024 | 2024-06-12 |
Meta builds two 24K clusters (one RoCE-optimized, one InfiniBand) and relies on PyTorch to move research to production, mapping tensor/pipeline/data… | ||
| Scaling RoCE Networks for AI Training | Networking @Scale 2023 | 2023-09-07 |
Meta runs RoCEV2 on commodity Ethernet to support AI training fabrics with tens of thousands of GPUs, using servers with 8 GPUs and 8 NICs and 16 uplinks per… | ||
| nothing matches | ||