VR2R-0224_r1

VR2R-0224

① SA Source

• Source: 開啟完整 SA 文章
• Section: BlueField-4
• Line hint: 194

Context Before

KV cache grows linearly with sequence length and multiplicatively with workload parallelism, quickly expanding beyond what any single tier of memory was designed to hold.

Evidence

GPU HBM, while unmatched in bandwidth and latency, is not enough on its own to store KV especially for longer sequence length queries that are becoming popular between turns or tool calls.

Context After

Host DRAM extends capacity but remains node-bound and limited in aggregate footprint and ultimately has limited capacity.

② Atomic Claim

GPU HBM 雖然在 bandwidth 與 latency 上無可匹敵，但單靠 HBM 不足以儲存 KV，尤其是愈來愈常見、跨 turns 或 tool calls 的較長 sequence-length queries。

• Epistemic Mode: ASSERTED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/GPU",
        "label": "GPU"
      },
      "role": "compute_device"
    },
    {
      "node": {
        "id": "04_knowledge_base/HBM",
        "label": "HBM"
      },
      "role": "memory"
    }
  ],
  "qualifiers": {
    "condition_text": "HBM alone is insufficient for KV storage especially for longer-sequence queries",
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "GPU_HBM_KV_CAPACITY_LIMIT"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
compute_device	GPU	GPU
memory	HBM	HBM

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。