clm_c7fb7760117142a582996d0ff17af46a_r1

VR2-0607

① SA Source

• Source: 開啟完整 SA 文章
• Section: GB200 Hardware Architecture - Component Supply Chain & BOM
• Line hint: 733

Context Before

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

While the VR NVL72 system features GPUs and scale-up switches that are connected by copper cables, the VR HGX system features servers consisting of eight Rubin GPUs and four NVLink Switch chips. The second meaningful difference between the NVL72 and HGX deployments is that the former has a scale-out bandwidth of 1.6T per GPU while the latter only has a scale-out bandwidth of 800G per GPU. How is it that all Rubin 200 deployments use CX-9 NICs even though some deployments have half the per GPU scale-out bandwidth?

Evidence

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

Context After

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

HGX Rubin NVL8 server 由 8 個 800G CX-9 NIC packages 組成，每顆 GPU 對應 1 個 NIC，因此 scale-out bandwidth 並未高於前代 HGX B300 server。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "comparison_expression": "HGX Rubin NVL8 server 由 8 個 800G CX-9 NIC packages 組成，每顆 GPU 對應 1 個 NIC，因此 scale-out bandwidth 並未高於前代 HGX B300 server。",
  "entities": [
    {
      "id": "04_knowledge_base/HGX",
      "label": "HGX"
    },
    {
      "id": "04_knowledge_base/Vera Rubin",
      "label": "Rubin"
    },
    {
      "id": "04_knowledge_base/NVL8",
      "label": "NVL8"
    },
    {
      "id": "04_knowledge_base/800G",
      "label": "800G"
    },
    {
      "id": "04_knowledge_base/NIC",
      "label": "NIC"
    },
    {
      "id": "04_knowledge_base/GPU",
      "label": "GPU"
    },
    {
      "id": "04_knowledge_base/NVIDIA B300",
      "label": "B300"
    }
  ],
  "frame_type": "COMPARISON",
  "metric": "BANDWIDTH",
  "operator": "GREATER_THAN",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [
      "8",
      "800",
      "9",
      "1"
    ],
    "temporal_mentions": []
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
comparison_entity_0	HGX	HGX
comparison_entity_1	Rubin	04_knowledge_base/Vera Rubin
comparison_entity_2	NVL8	NVL8
comparison_entity_3	800G	800G
comparison_entity_4	NIC	NIC
comparison_entity_5	GPU	GPU
comparison_entity_6	B300	04_knowledge_base/NVIDIA B300

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。