clm_9bad6ad08d6442debb140b5c3c5c2abe_r1

VR2-0683

① SA Source

• Source: 開啟完整 SA 文章
• Section: Ethernet-Based Cluster Deployments
• Line hint: 853

Context Before

Recall from above that a Vera Rubin NVL72 cluster built with Q3400-X800 switches at 1.6T logical ports per GPU cannot scale beyond the maximum cluster size of 93,312 GPUs. Even if future switch generations continue to double the maximum possible switching capacity per switch box, the per GPU bandwidth is also expected to double, which means that the effective logical port count in a cluster network is unlikely to change.

What this means is that deployment of large-scale GPU clusters will continue to require multi-plane network architectures. Do note however that multi-plane networking architectures are not limited by size and we have also seen NVIDIA reference architecture deployments with multi-plane clusters that are significantly below 100k GPUs each.

Evidence

In multi-plane network architectures using the SN6600 switch instead of the CPO switch, each GPU fans out to multiple switch boxes using fiber shuffles

Context After

Some of Nvidia’s CPO SKUs – such as the SN6800 and Q3450 –contain such a fiber shuffle within the switch box, with each optical engine fanning out to different logical ports. They therefore deliver higher aggregate bandwidth – 409.6T and 115.2T respectively than is possible with a switch box based on a single Switch ASIC.

② Atomic Claim

使用 SN6600 switch、而非 CPO switch 的 multi-plane network architecture 中，每顆 GPU 透過 fiber shuffle 向多個 switch boxes fan out。

• Epistemic Mode: ASSERTED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/Spectrum 6600",
        "label": "SN6600"
      },
      "role": "user_or_subject"
    },
    {
      "node": {
        "id": "04_knowledge_base/CPO",
        "label": "CPO"
      },
      "role": "used_entity"
    },
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "used_entity"
    },
    {
      "node": {
        "id": "04_knowledge_base/Fiber interconnect",
        "label": "fiber"
      },
      "role": "used_entity"
    }
  ],
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "USES"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
user_or_subject	SN6600	04_knowledge_base/Spectrum 6600
used_entity	CPO	CPO
used_entity	GPU	GPU
used_entity	fiber	04_knowledge_base/Fiber interconnect

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。