IX2-0078
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Disaggregated Prefill - Line hint:
109
Context Before
When prefill and decode performed on the same engine, prefill constantly disrupts decode batches leading to worse overall performance.
Disaggregated Prefill
Evidence
Context After
Tensor Parallel, Expert Parallel, Data Parallel (TP, EP, DP)
TP allows for maximize interactivity at small batch sizes, but it must carry out an all-reduce at every layer. EP shards experts, exploiting MoE sparsity, with the drawback being an all-to-all collective (which is more costly than simpler collectives like all-reduce) is carried out for MoE layers and can be imbalanced at small batches. DP replicates the entire model (or just parts of a model, like attention) on multiple groups of GPUs (ranks) and then load balances requests among ranks. It is the simplest to scale, but repeats weight loading which can be wasteful at scale.
② Atomic Claim
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"attribute": "UNSPECIFIED_ATTRIBUTE",
"context_nodes": [
{
"id": "04_knowledge_base/Decode",
"label": "decode"
}
],
"entity": {
"id": "04_knowledge_base/Prefill",
"label": "prefill"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"value": {
"numeric_mentions": [],
"value_text": "分離後的 prefill 與 decode pools 可以各自獨立調校，並依 workload 需求分別擴展。"
}
}④ Canonical Entity Mapping
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。