VR2-0272
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
For GB200, the GPU does not have direct access to the ConnectX-7
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 GB200 中，GPU 無法直接存取 ConnectX-7。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"frame_type": "NARY_RELATION",
"participants": [
{
"node": {
"id": "04_knowledge_base/GB200",
"label": "GB200"
},
"role": "endpoint_or_system"
},
{
"node": {
"id": "04_knowledge_base/GPU",
"label": "GPU"
},
"role": "connected_entity"
},
{
"node": {
"id": "04_knowledge_base/ConnectX-7",
"label": "ConnectX-7"
},
"role": "connected_entity"
}
],
"qualifiers": {
"condition_text": null,
"numeric_mentions": [
"7"
],
"temporal_mentions": []
},
"relation_type": "CONNECTS_TO"
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| endpoint_or_system | GB200 | GB200 |
| connected_entity | GPU | GPU |
| connected_entity | ConnectX-7 | ConnectX-7 |
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。