IX2-0161
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Nvidia versus AMD Disagg Prefill - Line hint:
266
Context Before
Source: SemiAnalysis InferenceX ↗
Evidence
Note that when looking at output token only throughput for disaggregated inference configurations, we normalize throughout by the number of decode GPUs, not total GPUs
Context After
SemiAnalysis is free open source software and reader-supported. To receive new posts and support our work, consider becoming a free or paid subscriber.
Subscribed
② Atomic Claim
分析 disaggregated inference 的 output-token-only throughput 時，InferenceX 是以 decode GPUs 數量做正規化，而非全部 GPUs 數量。
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"attribute": "THROUGHPUT",
"context_nodes": [
{
"id": "04_knowledge_base/Decode",
"label": "decode"
},
{
"id": "04_knowledge_base/GPU",
"label": "GPUs"
}
],
"entity": {
"id": "04_knowledge_base/Disaggregated inference",
"label": "disaggregated inference"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"value": {
"numeric_mentions": [],
"value_text": "分析 disaggregated inference 的 output-token-only throughput 時，InferenceX 是以 decode GPUs 數量做正規化，而非全部 GPUs 數量。"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| entity | disaggregated inference | 04_knowledge_base/Disaggregated inference |
| context_0 | decode | Decode |
| context_1 | GPUs | 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。