2025-12-04_aws-trainium3-deep-dive-a-potential::TRN3-0020
① SA Source
- Source: 開啟完整 SA 文章
- Section:
JBOG/ Compute Tray - Line hint:
277
Context Before
Source: SemiAnalysis
The Trainium3 NL32x2 Switched is the time to market SKU for Trainium3 that supports a switched scale-up architecture. As it is an air-cooled rack, rack power density remains low. It can be thought of as the same power density profile as the Trainium3 NL32x2 3D Torus with the main difference simply being the addition of scale-up switch trays. Trainium3 NL32x2 Switched is also the only SKU with a switched scale-up architecture that can be deployed in datacenters that are not liquid cooled ready.
Evidence
Context After
Trainium3 NL72x2 Switched (Teton3 Max)
Both the Trainium3 NL32x2 Switched and Trainium3 NL72x2 Switched use an all-to-all switched architecture, but the Trainium3 NL72x2 Switched is the rack architecture that is most comparable to Nvidia’s GB200 NVL72 Oberon architecture. Other than both Oberon and Trainium3 NL72x2 Switched using liquid cooling, Trainium3 NL72x2 Switched integrates the CPU into the compute trays just like Nvidia does with Grace and Vera on the same compute tray as the GPUs. By comparison, Trainium NL32x2 Switched still uses a disaggregated CPU node. Like Oberon, Trainium NL72x2 Switched uses cold plates for liquid cooling of the Trainium3 accelerators and the Graviton 4 CPUs. The big difference of Trainium NL72x2 Switched from the Oberon architecture is the use of cross-rack connectivity to increase the scale up world size to span over two racks.
② Atomic Claim
SemiAnalysis 預期 2026 年多數 Trainium3 出貨將採 NL32x2 Switched SKU。
- Epistemic Mode:
EXPECTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"attribute": "2026_majority_rack_sku",
"context_nodes": [],
"entity": {
"id": "04_knowledge_base/Trainium",
"label": "Trainium3"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [
"2026"
],
"temporal_mentions": []
},
"value": {
"numeric_mentions": [
"2026"
],
"value_text": "NL32x2 Switched"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| entity | Trainium3 | Trainium |
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。