VR2-0387

① SA Source

Context Before

Our VR NVL72 Component BoM and Power Budget Model has provides the $ content breakdown of high end CCL and PCB content for VR NVL72 vs GB200/GB300.

Compute Tray: Thermal

Evidence

GB200 and GB300 compute trays adopt a hybrid of 85% liquid cooling and 15% air cooling

Context After

image

Source: Nvidia VR NVL72 Component BoM and Power Budget Model

② Atomic Claim

GB200GB300 compute trays 採 85% liquid cooling+15% air cooling 的 hybrid design。

  • Epistemic Mode: ASSERTED
  • Mapping Status: COMPLETE

③ Semantic Frame

{
  "attribute": "COMPUTE_PERFORMANCE",
  "context_nodes": [
    {
      "id": "04_knowledge_base/GB300",
      "label": "GB300"
    },
    {
      "id": "04_knowledge_base/液冷散熱",
      "label": "liquid cooling"
    },
    {
      "id": "04_knowledge_base/Air cooling",
      "label": "air cooling"
    }
  ],
  "entity": {
    "id": "04_knowledge_base/GB200",
    "label": "GB200"
  },
  "frame_type": "ATTRIBUTE",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [
      "85%",
      "15%"
    ],
    "temporal_mentions": []
  },
  "value": {
    "numeric_mentions": [
      "85%",
      "15%"
    ],
    "value_text": "GB200 與 GB300 compute trays 採 85% liquid cooling+15% air cooling 的 hybrid design。"
  }
}

④ Canonical Entity Mapping

RoleSurface LabelCanonical Target
entityGB200GB200
context_0GB300GB300
context_1liquid cooling04_knowledge_base/液冷散熱
context_2air cooling04_knowledge_base/Air cooling

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。