SA260324R-0032
① SA Source
- Source: 開啟完整 SA 文章
- Section:
LPU chip - Line hint:
43
Context Before
Concretely, LPU architecture has VXM slices for vector operations, MEM slices for loading/storing data, SXM slices for tensor shape manipulation, and MXM slices for performing matrix multiplication. Spatially, the slices are laid out horizontally, allowing the data to stream horizontally. Within a slice, instructions are pumped vertically across units.
Evidence
Conceptually, LPU resembles a systolic array that pumps instructions vertically and data horizontally.
Context After
The data flow and instruction flow design requires fine-grained pipelining to achieve high performance. Since LPU architecture makes computation deterministic, the compiler can aggressively schedule and overlap instructions to hide latency. The LPU’s use of high bandwidth SRAM and aggressive pipelining are the two main factors that enable LPU’s low latency.
② Atomic Claim
概念上，LPU 類似 systolic array：instructions 沿垂直方向傳送，而資料沿水平方向流動。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"additional_nodes": [],
"frame_type": "RELATION",
"object": {
"id": "04_knowledge_base/Systolic Array",
"label": "systolic array"
},
"predicate": "RESEMBLES",
"qualifiers": {
"condition_text": "instructions vertically；data horizontally",
"numeric_mentions": [],
"temporal_mentions": []
},
"subject": {
"id": "04_knowledge_base/Groq LPU",
"label": "LPU"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| subject | LPU | 04_knowledge_base/Groq LPU |
| object | systolic array | 04_knowledge_base/Systolic Array |
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。