IX2-0240
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Nvidia Disagg Prefill and WideEP - Line hint:
390
Context Before
At higher concurrencies, this tradeoff changes. Expert activation becomes more evenly distributed across larger batch sizes, and EP’s communication and weight-loading advantages dominate over TP’s expensive per-layer all-reduce. In the middle of the curve, hybrid TP+EP configurations balance both concerns using small TP groups within each expert for load balancing while EP is used across the wider set of GPUs to amortize weights and reduce communication.
For higher interactivity levels (low batch size), large scale-up world sizes tend not to deliver stronger performance. B300 disagg over IB has the same performance as GB300 with NVL72, since the workload is latency-bound, not bandwidth-bound. The massive NVLink bandwidth advantage of NVL72 doesn’t matter because not even the much slower IB link is saturated by the tiny batches of tokens in flight.
Evidence
Separating them onto dedicated GPU pools lets each run a workload matched to its characteristics, improving effective utilization
Context After
Side Note: the 10x inference engineers at TogetherAI noticed an pattern for multi-turn traffic where the requirements of first turn prefill is much different from the following turns prefill’s and disaggregrated it leading to better TTFT performance. ↗
② Atomic Claim
將兩者分配到專用 GPU pools 後，每個 pool 都能執行更符合自身特性的 workload，進而提高有效利用率。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"attribute": "UTILIZATION",
"context_nodes": [],
"entity": {
"id": "04_knowledge_base/GPU",
"label": "GPU"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"value": {
"numeric_mentions": [],
"value_text": "將兩者分配到專用 GPU pools 後，每個 pool 都能執行更符合自身特性的 workload，進而提高有效利用率。"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| entity | GPU | GPU |
⑤ Human Review
請在 Properties 逐項確認：
- 原文 → Atomic Claim 是否忠實
- Atomic Claim → Semantic Frame 是否忠實
- Canonical Entity mapping 是否正確
- Epistemic mode 是否保留原文語氣
- 最後選擇
review_action
Review state
Markdown 內文不是正式 approval。只有 Apply bridge 寫入的 Decision Ledger event 才是正式決策。