clm_13b3e6f8a10d4051a0e3559caea48409_r2

VR2R-0590

回到這篇 SA Coverage

① SA Source

• Source: 開啟完整 SA 文章
• Section: Connecting GPUs in the Scale-Out InfiniBand Network
• Line hint: 753

Context Before

Source: SemiAnalysis AI Networking Model ↗

Evidence

The InfiniBand architecture for the HGX Rubin NVL8 server is therefore effectively a single-plane, 8-rail network consisting of one 800G uplink from each HGX Rubin NVL8 GPU to a leaf switch.

Context After

Source: SemiAnalysis AI Networking Model ↗

② Atomic Claim

因此，HGX Rubin NVL8 server 的 InfiniBand architecture，實際上相當於 single-plane、8-rail network；每顆 HGX Rubin NVL8 GPU 各有一條 800G uplink 連到 leaf switch。

• Epistemic Mode: INFERRED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/InfiniBand",
        "label": "InfiniBand"
      },
      "role": "fabric"
    },
    {
      "node": {
        "id": "04_knowledge_base/HGX",
        "label": "HGX"
      },
      "role": "platform"
    },
    {
      "node": {
        "id": "04_knowledge_base/Vera Rubin",
        "label": "Rubin"
      },
      "role": "processor"
    },
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "processor_type"
    }
  ],
  "qualifiers": {
    "condition_text": "inferred architecture; leaf switch is plain text",
    "numeric_mentions": [
      "1",
      "8",
      "800G"
    ],
    "temporal_mentions": []
  },
  "relation_type": "HGX_RUBIN_NVL8_INFINIBAND_SINGLE_PLANE_EIGHT_RAIL"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
fabric	InfiniBand	InfiniBand
platform	HGX	HGX
processor	Rubin	04_knowledge_base/Vera Rubin
processor_type	GPU	GPU

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

因此，HGX Rubin NVL8 server 的 InfiniBand architecture，實際上相當於 single-plane、8-rail network；每顆 HGX Rubin NVL8 GPU 各有一條 800G uplink 連到 leaf switch。

V2 Semantic Frame Draft

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/InfiniBand",
        "label": "InfiniBand"
      },
      "role": "fabric"
    },
    {
      "node": {
        "id": "04_knowledge_base/HGX",
        "label": "HGX"
      },
      "role": "platform"
    },
    {
      "node": {
        "id": "04_knowledge_base/Vera Rubin",
        "label": "Rubin"
      },
      "role": "processor"
    },
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "processor_type"
    }
  ],
  "qualifiers": {
    "condition_text": "inferred architecture; leaf switch is plain text",
    "numeric_mentions": [
      "1",
      "8",
      "800G"
    ],
    "temporal_mentions": []
  },
  "relation_type": "HGX_RUBIN_NVL8_INFINIBAND_SINGLE_PLANE_EIGHT_RAIL"
}

V2 Revision Metadata Draft

{
  "epistemic_mode": "INFERRED",
  "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。