Metraipolis › archive › dc_3bef757ca827
Divide and conquer: Scalable performance and energy in MCM GPUs — arXiv:2610.03061
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03061
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:53:46.213Z
- SHA-256 of the archived text
7aa7f6fe155a23d4a59bd558cc684924874beca897a45262d10e810275e70a34- Stable URL of this page
https://metraipolis.com/documento/dc_3bef757ca827
How to cite
Divide and conquer: Scalable performance and energy in MCM GPUs — arXiv:2610.03061 — arXiv, https://arxiv.org/abs/2610.03061 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3bef757ca827 (Metraipolis archive dc_3bef757ca827, sha256 7aa7f6fe155a23d4a59bd558cc684924874beca897a45262d10e810275e70a34)
BibTeX
@misc{dc3bef757ca827,
title = {Divide and conquer: Scalable performance and energy in MCM GPUs — arXiv:2610.03061},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03061},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3bef757ca827; sha256 7aa7f6fe155a23d4a59bd558cc684924874beca897a45262d10e810275e70a34},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03061. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-chip-module (MCM) GPUs offer a promising path to scale compute capability beyond monolithic designs by integrating multiple chiplets on a common package. However, the impact of disaggregation on performance scalability and energy consumption remains underexplored. The design space grows rapidly across dimensions such as SMs per chiplet, chiplet count, and interconnection network. The inter-chiplet network is particularly critical, as it determines whether additional compute resources translate into performance gains. This limited understanding leaves industry and research without clear guidance on the performance and energy trade-offs of MCM GPU scaling. In this work, we investigate whether distributing compute and memory capability across multiple chiplets offers a more scalable alternative to concentrating resources. We quantify their effects on performance, energy, and efficiency and examine how inter-chiplet topology influences scalability at different system sizes. Our results demonstrate that a 16 chiplet Torus configuration with 256 SMs delivers a remarkable $2.40\times$ performance improvement over a state-of-the-art MCM architecture with the same compute capability, while simultaneously reducing energy consumption by $4.45\times$. These substantial gains provide evidence that disaggregation is a first-order architectural factor and will be critical to unlocking the performance and energy-efficiency potential of next-generation GPUs. 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/2610.03061 — 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 2026-10-05T22:46:40.544Z dalla pagina https://arxiv.org/list/cs.AR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).