Homa: The End of TCP for AI Clusters
signa11 · hn · 2026-10-05
An HN discussion around Stanford's Homa transport protocol, pitched as a replacement for TCP in AI clusters. Homa redesigns the transport layer for AI training traffic dominated by short messages and incast congestion, claiming major gains in tail latency over TCP and RDMA-based setups. The post links a talk video plus a Register writeup of the Stanford work.
More from Infra
- Ex-Waymo exec Abe Murray: CUDA and ISAs won't stay moats, huge chip innovation incoming — IgorCarron · 2026-11-20
- MIT and NVIDIA's QDA uses agents to auto-design LLM quantization recipes, beating top public ones by up to 2.2 points — omarsar0 · 2026-11-20
- Caixin: US remote-access cutoff and Nvidia CEO exit leave China's tech giants halting overseas compute deals — Gregory_C_Allen · 2026-11-20
- QuixiInfer: Rust Engine Runs 125B MoE Model on a Single RTX 5090 — QuixiAI · 2026-11-20
- QuixiInfer: from-scratch Rust engine runs DeepSeek with 256k context on a single RTX 5090 at ~30 tok/s — QuixiAI · 2026-11-20
- AMD's Lisa Su: data center to hit ~80% of revenue by 2027, sees $220B server CPU market — AccBalanced · 2026-11-20