clm_de6da2d49bd44e4e96d4f958f090a2f5_r1

IX2-0402

① SA Source

• Source: 開啟完整 SA 文章
• Section: WideEP
• Line hint: 636

Context Before

Consider serving DeepSeek R1 on a single 8-GPU server. At 671B parameters, some form of parallelism is required to fit the model across available HBM. The naive approach is tensor parallelism (TP), which shards every weight matrix across all GPUs. This works well for dense models but ignores the sparse activation pattern of MoE. With TP=8, each expert’s weights are sharded across all 8 GPUs, meaning every expert activation requires an all-reduce across all GPUs & the reduction dims of the GEMM is smaller leading to lower arithmetic intensity, even though only 8 of 256 experts activate per token. TP treats each expert like a dense layer, paying full cross-GPU communication cost while the model’s sparsity goes unexploited.

Expert parallelism takes a more well-suited approach, assigning whole experts to individual GPUs. With EP=8, we divide the 256 experts per layer across 8 GPUs for a total of 32 experts/layer/GPU. Each GPU holds approximately 1/8th of the expert weights plus a full replica of the non-expert weights (attention projections, embeddings, normalization, and the shared expert). Since roughly 90%+ of DeepSeek R1’s parameters are routed expert weights, EP captures most of the memory savings, and replicating the remaining less than 30B non-expert parameters across all 8 GPUs is affordable.

Evidence

tokens are dispatched to the appropriate GPUs via all-to-all communication

Context After

_An EP8 DP8 deployment of DeepSeek R1. All 256 experts per layer are divided evenly among the 8 GPUs, whereas attention along with other non-expert weights (shared expert, gating network, RMSNorm, LM head, etc.) are replicated across all 8 DP ranks. _Source: SemiAnalysis

② Atomic Claim

接著 tokens 透過 all-to-all communication 被送往對應的 GPUs。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "attribute": "UNSPECIFIED_ATTRIBUTE",
  "context_nodes": [],
  "entity": {
    "id": "04_knowledge_base/GPU",
    "label": "GPUs"
  },
  "frame_type": "ATTRIBUTE",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "value": {
    "numeric_mentions": [],
    "value_text": "接著 tokens 透過 all-to-all communication 被送往對應的 GPUs。"
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
entity	GPUs	GPU

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。