SA260324R-0119
① SA Source
- Source: 開啟完整 SA 文章
- Section:
GPU and LPU Integration: Attention FFN Disaggregation (AFD) - Line hint:
113
Context Before
Source: SemiAnalysis, MegaScale-Infer
Evidence
With AFD, token routing from GPUs to LPUs can become the bottleneck, especially under strict latency constraints. The token routing flow involves two operations: dispatch and combine. In the dispatch step, we route each token to their top k experts with an All-to-All collective operation. After experts complete their computation, we perform the combine step, where the outputs are sent back to the source location with a reverse All-to-All collective, continuing the next layer’s computation.
Context After
Source: SemiAnalysis
② Atomic Claim
expert 完成運算後會進入 combine 階段，輸出透過 reverse All-to-All collective 傳回來源位置，接著繼續下一層運算。
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"frame_type": "CLAIM_ONLY",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"reason": "COMBINE_AND_ALL_TO_ALL_HAVE_NO_CANONICAL_ENTITIES",
"statement": "expert 完成運算後會進入 combine 階段，輸出透過 reverse All-to-All collective 傳回來源位置，接著繼續下一層運算。"
}④ Canonical Entity Mapping
- 此
CLAIM_ONLYrevision 沒有 canonical Entity mapping。
⑤ V2 Draft（需要修正時直接編輯）
Inline repair
只編輯以下三個 Draft 區塊。完成後將
review_action設為REQUEST_REPAIR，再手動執行 Repair Preview。
執行腳本前，Draft 不會寫入 Claim Store 或 Graph。
V2 Atomic Claim Draft
expert 完成運算後會進入 combine 階段，輸出透過 reverse All-to-All collective 傳回來源位置，接著繼續下一層運算。
V2 Semantic Frame Draft
{
"frame_type": "CLAIM_ONLY",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"reason": "COMBINE_AND_ALL_TO_ALL_HAVE_NO_CANONICAL_ENTITIES",
"statement": "expert 完成運算後會進入 combine 階段，輸出透過 reverse All-to-All collective 傳回來源位置，接著繼續下一層運算。"
}V2 Revision Metadata Draft
{
"epistemic_mode": "ASSERTED",
"mapping_status": "PARTIAL"
}⑥ 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。