Metraipolis › archive › dc_99b0ee926fc0

TranSAGE: Automating Spatial Accelerator Generation for Large-Scale Transformer Models — arXiv:2610.32279

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.32279
License (as archived)
CC0-1.0
Archived
2026-10-26T19:54:56.932Z
SHA-256 of the archived text
58608ebac657bdbd521e08504c1b9b09bee5dccda57f793d357525f8baaf0312
Stable URL of this page
https://metraipolis.com/documento/dc_99b0ee926fc0

How to cite

TranSAGE: Automating Spatial Accelerator Generation for Large-Scale Transformer Models — arXiv:2610.32279 — arXiv (en), https://arxiv.org/abs/2610.32279 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_99b0ee926fc0 (Metraipolis archive dc_99b0ee926fc0, sha256 58608ebac657bdbd521e08504c1b9b09bee5dccda57f793d357525f8baaf0312)
BibTeX
@misc{dc99b0ee926fc0,
  title = {TranSAGE: Automating Spatial Accelerator Generation for Large-Scale Transformer Models — arXiv:2610.32279},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.32279},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_99b0ee926fc0; sha256 58608ebac657bdbd521e08504c1b9b09bee5dccda57f793d357525f8baaf0312},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32279. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large-scale Transformers have achieved remarkable success across NLP, computer vision, and robotics, but their growing computational and memory demands challenge efficient inference. Spatial accelerators offer advantages in on-chip bandwidth and energy efficiency, yet designing them for large models remains costly and time-consuming. Existing accelerator generators typically focus on single-operation support, overlooking the benefits of multi-operation pipelining, which limits throughput and energy efficiency. We propose TranSAGE, a compiler-based framework that jointly optimizes scheduling for large-scale Transformers under multi-operation pipelining and generates corresponding accelerator implementations. Our approach integrates a design space exploration (DSE) engine with multi-operation dataflow optimization and a dynamic programming-based graph partitioning algorithm to efficiently identify pipeline-friendly operation clusters and their optimal dataflow parameters. We further design spatial architecture primitives and lowering passes to automatically generate efficient RTL implementations. Evaluations on four representative LLMs demonstrate that TranSAGE achieves up to 2.41x higher throughput and 13.04x higher energy efficiency than commercial GPU/NPU baselines. Compared with state-of-the-art accelerator generators, TranSAGE achieves up to 11.33x higher throughput and 6.07x higher energy efficiency. Our code is available at this https URL .

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.AR, nuovi lavori del giorno).
Categorie arXiv: Hardware Architecture (cs.AR).
Pagina del lavoro: https://arxiv.org/abs/2610.32279 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-26T19:46:18.292Z dalla pagina https://arxiv.org/list/cs.AR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).