VR2-0544
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Networking: NVLink 6, Rubin Scale Out - Line hint:
635
Context Before
We have more detailed power and architecture details in our VR NVL72 Component BoM and Power Budget Model ↗.
Networking: NVLink 6, Rubin Scale Out
Evidence
Context After
The below table shows the evolution of scale-up and scale-out networking speeds. NVLink 6 used in Vera Rubin doubles NVLink bandwidth by implementing bi-directional signaling over the same number of copper cables - effectively delivering 4 Lanes of 200G per NVLink. Much more on this in the following sections.
② Atomic Claim
隨 GPU scale-up 與 scale-out bandwidth 約每 18 個月加倍，NVIDIA rack 中的 copper infrastructure 也持續創新，以支援更高 bandwidth workload。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"comparison_expression": "隨 GPU scale-up 與 scale-out bandwidth 約每 18 個月加倍，NVIDIA rack 中的 copper infrastructure 也持續創新，以支援更高 bandwidth workload。",
"entities": [
{
"id": "04_knowledge_base/GPU",
"label": "GPU"
},
{
"id": "02_companies/NVDA",
"label": "NVIDIA"
},
{
"id": "04_knowledge_base/Copper",
"label": "copper"
}
],
"frame_type": "COMPARISON",
"metric": "BANDWIDTH",
"operator": "MULTIPLE_OF",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [
"18"
],
"temporal_mentions": []
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| comparison_entity_0 | GPU | GPU |
| comparison_entity_1 | NVIDIA | NVDA |
| comparison_entity_2 | copper | Copper |
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。