clm_e1402ecfae91425cac61731bd6a21c00_r3

VR2R-0261

回到這篇 SA Coverage

① SA Source

• Source: 開啟完整 SA 文章
• Section: Compute Tray Topology
• Line hint: 369

Context Before

Source: Nvidia VR NVL72 BoM and Power Budget Model ↗

Evidence

Firstly, the connection between the GPU and the ConnectX NICs evolved from GB200 to GB300 then to Vera Rubin. For GB200, the GPU does not have direct access to the ConnectX-7. Instead, B200 connects to Grace CPU via C2C connection then the Grace CPU talks to the ConnectX-7 over PCIe 5. For GB300, Nvidia introduced NIC direct to ConnectX-8, which allows the B300 GPU to communicate directly with the ConnectX-8 NIC without going through the Grace CPU.

Context After

Essentially, this means ConnectX-8 has two hosts, Grace CPU and B300 GPU. This improves latency in the backend network. However, for VR NVL72, the direct connection between Rubin GPU and ConnectX-9 is reverted back to the same design as GB200, as Rubin does not have PCIe bandwidth for two Connect-9. Rubin connects to Vera via C2C link, then Vera will connect to ConnectX-9 via PCIe6 lanes.

Secondly, the local NVMe storage for Rubin has been moved to a different location from that of NVMe storage in Grace Blackwell. Previously, local NVMe storage was managed by BlueField-3. For VR NVL72, the local NVMe storage is physically on the Orchid module managed by the ConnectX-9.

② Atomic Claim

在 GB300 中，Nvidia 導入 NIC direct to ConnectX-8，使 B300 GPU 可直接與 ConnectX-8 NIC 通訊，不需經過 Grace CPU。

• Epistemic Mode: ASSERTED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/GB300",
        "label": "GB300"
      },
      "role": "system"
    },
    {
      "node": {
        "id": "02_companies/NVDA",
        "label": "Nvidia"
      },
      "role": "vendor"
    },
    {
      "node": {
        "id": "04_knowledge_base/NVIDIA B300",
        "label": "B300"
      },
      "role": "accelerator"
    },
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "accelerator_type"
    },
    {
      "node": {
        "id": "04_knowledge_base/ConnectX-8",
        "label": "ConnectX-8"
      },
      "role": "network_device"
    },
    {
      "node": {
        "id": "04_knowledge_base/NIC",
        "label": "NIC"
      },
      "role": "network_device_type"
    },
    {
      "node": {
        "id": "04_knowledge_base/NVIDIA Grace",
        "label": "Grace CPU"
      },
      "role": "bypassed_cpu"
    }
  ],
  "qualifiers": {
    "condition_text": "direct GPU-to-NIC path",
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "B300_DIRECT_CONNECTX8_WITHOUT_GRACE"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
system	GB300	GB300
vendor	Nvidia	NVDA
accelerator	B300	04_knowledge_base/NVIDIA B300
accelerator_type	GPU	GPU
network_device	ConnectX-8	ConnectX-8
network_device_type	NIC	NIC
bypassed_cpu	Grace CPU	04_knowledge_base/NVIDIA Grace

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

在 GB300 中，Nvidia 導入 NIC direct to ConnectX-8，使 B300 GPU 可直接與 ConnectX-8 NIC 通訊，不需經過 Grace CPU。

V2 Semantic Frame Draft

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/GB300",
        "label": "GB300"
      },
      "role": "system"
    },
    {
      "node": {
        "id": "02_companies/NVDA",
        "label": "Nvidia"
      },
      "role": "vendor"
    },
    {
      "node": {
        "id": "04_knowledge_base/NVIDIA B300",
        "label": "B300"
      },
      "role": "accelerator"
    },
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "accelerator_type"
    },
    {
      "node": {
        "id": "04_knowledge_base/ConnectX-8",
        "label": "ConnectX-8"
      },
      "role": "network_device"
    },
    {
      "node": {
        "id": "04_knowledge_base/NIC",
        "label": "NIC"
      },
      "role": "network_device_type"
    },
    {
      "node": {
        "id": "04_knowledge_base/NVIDIA Grace",
        "label": "Grace CPU"
      },
      "role": "bypassed_cpu"
    }
  ],
  "qualifiers": {
    "condition_text": "direct GPU-to-NIC path",
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "B300_DIRECT_CONNECTX8_WITHOUT_GRACE"
}

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。