{
  "schema_version": "1.1",
  "id": "s8:https://www.youtube.com/watch?v=eZ8WWZzoaR0",
  "slug": "homa-the-end-of-tcp-for-ai-clusters-john-ousterhout-stanford-1idp8cs",
  "url": "https://feed7.dev/p/homa-the-end-of-tcp-for-ai-clusters-john-ousterhout-stanford-1idp8cs",
  "title": "Homa: The End of TCP for AI Clusters — John Ousterhout, Stanford",
  "why_included": "Homa targets the small-message tail latency that can idle GPUs in distributed agentic inference, using message boundaries, short-message priority, and receiver-led congestion control.",
  "summary": "Ousterhout argues that AI traffic is shifting from giant transfers toward latency-sensitive synchronization. In his sample benchmark, short-message P99 was over 1 ms with TCP and under 100 μs with Homa, about **13x faster**; long messages were nearly **2x faster**.",
  "practical_implication": "If distributed inference stalls between millisecond-scale compute phases, measure small-message tail latency rather than aggregate throughput alone. Homa's message-aware RPCs, shortest-remaining-time priority, and receiver-issued grants directly target queueing behind large transfers.",
  "agent_context": "Ousterhout argues that AI traffic is shifting from giant transfers toward latency-sensitive synchronization. In his sample benchmark, short-message P99 was over 1 ms with TCP and under 100 μs with Homa, about **13x faster**; long messages were nearly **2x faster**.\n\nIf distributed inference stalls between millisecond-scale compute phases, measure small-message tail latency rather than aggregate throughput alone. Homa's message-aware RPCs, shortest-remaining-time priority, and receiver-issued grants directly target queueing behind large transfers.\n\nThese figures come from **one sample benchmark**, not a production comparison across varied clusters. Homa is available as a Linux kernel module and is being worked toward upstream inclusion, so adoption carries kernel and deployment risk.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=eZ8WWZzoaR0",
    "published_at": "2026-09-17T13:00:08.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "infra",
  "domains": [],
  "topics": [
    "agent-reliability"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "These figures come from **one sample benchmark**, not a production comparison across varied clusters. Homa is available as a Linux kernel module and is being worked toward upstream inclusion, so adoption carries kernel and deployment risk."
  ],
  "connected_context": {
    "meaning": "Homa narrows agent-infrastructure performance work to the network synchronization layer: when distributed inference alternates short compute phases with messages, small-message tail latency may matter more than bulk throughput. The sample results justify measuring that bottleneck, not assuming a production win; the protocol’s kernel module and incomplete upstream path make deployment maturity and operational risk part of any evaluation.",
    "corpus_size": 812,
    "generated_at": "2026-09-19T09:05:18.698Z",
    "connections": [
      {
        "title": "TokTier: Exact Stateful Tokenization for Agentic LLM Serving",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2607.29678v1",
        "feed7_url": "https://feed7.dev/p/2607-29678v1-1fuv1qw",
        "reason": "TokTier identifies tokenization as another hidden contributor to time-to-first-token, so evaluating Homa requires stage-level telemetry that separates network tail latency from front-end prompt processing."
      },
      {
        "title": "AI Gateway logs now have a dedicated page",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/ai-gateway-logs",
        "feed7_url": "https://feed7.dev/p/ai-gateway-logs-1272t5j",
        "reason": "Gateway request logs reinforce the need for per-request latency evidence, but Homa requires finer network-level tail measurements than aggregate gateway observations alone provide."
      },
      {
        "title": "How Web Data Infrastructure Powers the Next Generation of AI — Patricija Žemaitytė, Oxylabs",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=1UmZHb_E_SM",
        "feed7_url": "https://feed7.dev/p/how-web-data-infrastructure-powers-the-next-generation-of-ai-patricija-z-1ta4xzh",
        "reason": "The web-data talk similarly warns that development latency does not establish production resilience, supporting Homa’s need for representative multi-cluster testing beyond one sample benchmark."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-09-17T13:00:08.000Z",
  "modified_at": "2026-09-17T13:00:08.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/homa-the-end-of-tcp-for-ai-clusters-john-ousterhout-stanford-1idp8cs",
    "json": "https://feed7.dev/p/homa-the-end-of-tcp-for-ai-clusters-john-ousterhout-stanford-1idp8cs.json",
    "markdown": "https://feed7.dev/p/homa-the-end-of-tcp-for-ai-clusters-john-ousterhout-stanford-1idp8cs.md"
  }
}