clm_192ea385e2e34104a6d3f973fb8bce1c_r1

VR2-0606

① SA Source

• Source: 開啟完整 SA 文章
• Section: GB200 Hardware Architecture - Component Supply Chain & BOM
• Line hint: 731

Context Before

Vera Rubin NVL72 Scale-up Topology. Source: SemiAnalysis AI Networking Model ↗

Evidence

How is it that all Rubin 200 deployments use CX-9 NICs even though some deployments have half the per GPU scale-out bandwidth?

Context After

The HGX Rubin NVL8 server consists of eight 800G CX-9 NIC packages – one NIC per GPU – which means that the scale-out bandwidth does not increase from its predecessor, the HGX B300 server. The Vera Rubin NVL72 deployment on the other hand doubles the per GPU scale-out bandwidth to 1.6T, but not by doubling the bandwidth per NIC. Rather, the “1.6T NIC” attached to each Rubin chip is comprised of two 800G CX-9 packages that is connected to the Vera CPU by PCIe Gen 6.0 lanes.

Each compute tray on the VR NVL72 has eight 800G CX-9 NICs, but there are two possibilities for the number of OSFP cages - either one 1.6T OSFP cage per GPU for a total of 4 per compute tray, or two 800G OSFP cages per GPU for a total of 8 cages per compute tray. We think that the latter would be the more popular deployment assumption, and will be the base case for our discussion of scale-out networking architectures in later sections of the article.

② Atomic Claim

所有 Rubin 200 部署都使用 CX-9 NICs，但部分部署每顆 GPU scale-out bandwidth 只有一半。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "attribute": "BANDWIDTH",
  "context_nodes": [
    {
      "id": "04_knowledge_base/GPU",
      "label": "GPU"
    }
  ],
  "entity": {
    "id": "04_knowledge_base/Vera Rubin",
    "label": "Rubin"
  },
  "frame_type": "ATTRIBUTE",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [
      "200",
      "9"
    ],
    "temporal_mentions": []
  },
  "value": {
    "numeric_mentions": [
      "200",
      "9"
    ],
    "value_text": "所有 Rubin 200 部署都使用 CX-9 NICs，但部分部署每顆 GPU scale-out bandwidth 只有一半。"
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
entity	Rubin	04_knowledge_base/Vera Rubin
context_0	GPU	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。