IX2-0480
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Optimizing Inference with Wide EP + Disaggregated Serving - Line hint:
690
Context Before
Source: SemiAnalysis
Evidence
At this density, EP’s load imbalance becomes tolerable and its token-routing overhead is cheaper than the per-expert all-reduce required by TP
Context After
Source: SemiAnalysis InferenceX ↗
② Atomic Claim
到這個密度後，EP 的 load imbalance 已可接受，而 token-routing overhead 也低於 TP 對每個 expert 都需執行的 all-reduce 成本。
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"comparison_expression": "到這個密度後，EP 的 load imbalance 已可接受，而 token-routing overhead 也低於 TP 對每個 expert 都需執行的 all-reduce 成本。",
"entities": [
{
"id": "04_knowledge_base/All-Reduce",
"label": "all-reduce"
}
],
"frame_type": "COMPARISON",
"metric": "COST",
"operator": "LESS_THAN",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| comparison_entity_0 | all-reduce | All-Reduce |
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。