IX2-0449
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Disaggregated Prefill - Line hint:
664
Context Before
Disaggregated Prefill
Disaggregated prefill, sometimes referred to as prefill-decode (PD) disaggregation, is the process of performing prefill and decode phases of LLM inference on separate nodes. Prefill occurs when a request is first processed, and a forward pass is computed on all tokens at once, thereby “prefilling” the KV cache for this request. This is a compute-intensive operation as all tokens feed through the forward pass in parallel. Tokens are then generated or “decoded” one at a time, loading the KV cache from HBM at each decode step. This is a memory-intensive process as the growing KV cache is constantly being loaded.
Evidence
Chunked prefill mitigates this by breaking long prefills into smaller pieces
Context After
Source: DistServe ↗
② Atomic Claim
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"comparison_expression": "Chunked prefill 可把長 prefill 切成較小 pieces，減輕這個問題。",
"entities": [
{
"id": "04_knowledge_base/Prefill",
"label": "prefill"
}
],
"frame_type": "COMPARISON",
"metric": "COUNT",
"operator": "UNSPECIFIED_COMPARISON",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| comparison_entity_0 | prefill | Prefill |
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。