clm_8b3aca231db04212bf72994b25938a19_r1

SA260324R-0121::SPLIT02

回到這篇 SA Coverage

① SA Source

• Source: 開啟完整 SA 文章
• Section: GPU and LPU Integration: Attention FFN Disaggregation (AFD)
• Line hint: 119

Context Before

Source: SemiAnalysis

Evidence

To hide the communication latency of dispatch and combine, we employ ping pong pipeline parallelism. In addition to splitting batches into micro-batches and computation pipelining like standard pipeline parallelism, the tokens dispatched to the LPUs are combined back to the source GPUs, so they ping pong between the GPUs and the LPUs.

Context After

Source: MegaScale-Infer

② Atomic Claim

送到 LPU 的 token 會被 combine 回來源 GPU，因此 token 會在 GPU 與 LPU 之間來回 ping-pong。

• Epistemic Mode: ASSERTED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "source_endpoint"
    },
    {
      "node": {
        "id": "04_knowledge_base/Groq LPU",
        "label": "LPU"
      },
      "role": "dispatched_endpoint"
    },
    {
      "node": {
        "id": "04_knowledge_base/Pipeline parallelism",
        "label": "pipeline parallelism"
      },
      "role": "flow_mechanism"
    }
  ],
  "qualifiers": {
    "condition_text": "tokens dispatched to LPUs are combined back to the source GPU",
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "PING_PONG_TOKEN_FLOW"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
source_endpoint	GPU	GPU
dispatched_endpoint	LPU	04_knowledge_base/Groq LPU
flow_mechanism	pipeline parallelism	04_knowledge_base/Pipeline parallelism

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

送到 LPU 的 token 會被 combine 回來源 GPU，因此 token 會在 GPU 與 LPU 之間來回 ping-pong。

V2 Semantic Frame Draft

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "source_endpoint"
    },
    {
      "node": {
        "id": "04_knowledge_base/Groq LPU",
        "label": "LPU"
      },
      "role": "dispatched_endpoint"
    },
    {
      "node": {
        "id": "04_knowledge_base/Pipeline parallelism",
        "label": "pipeline parallelism"
      },
      "role": "flow_mechanism"
    }
  ],
  "qualifiers": {
    "condition_text": "tokens dispatched to LPUs are combined back to the source GPU",
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "PING_PONG_TOKEN_FLOW"
}

V2 Revision Metadata Draft

{
  "epistemic_mode": "ASSERTED",
  "mapping_status": "COMPLETE"
}

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。