Metraipolis › archive › dc_14cf7ead0bb4

Sustainable Processing-in-DRAM Integration and Replacement for Transformer Inference Servers — arXiv:2610.27345

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.27345
License (as archived)
CC0-1.0
Archived
2026-10-21T07:11:29.455Z
SHA-256 of the archived text
82c1ee62876a4a482abcc18c71fe68d5fb364231b07f18ae38bea9897a495f27
Stable URL of this page
https://metraipolis.com/documento/dc_14cf7ead0bb4

How to cite

Sustainable Processing-in-DRAM Integration and Replacement for Transformer Inference Servers — arXiv:2610.27345 — arXiv (en), https://arxiv.org/abs/2610.27345 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_14cf7ead0bb4 (Metraipolis archive dc_14cf7ead0bb4, sha256 82c1ee62876a4a482abcc18c71fe68d5fb364231b07f18ae38bea9897a495f27)
BibTeX
@misc{dc14cf7ead0bb4,
  title = {Sustainable Processing-in-DRAM Integration and Replacement for Transformer Inference Servers — arXiv:2610.27345},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27345},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_14cf7ead0bb4; sha256 82c1ee62876a4a482abcc18c71fe68d5fb364231b07f18ae38bea9897a495f27},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27345. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Processing-in-DRAM (PID) accelerators are emerging as a sustainable alternative to GPUs for transformer model inference. They bring compute to data, mitigate the memory bandwidth limitation, and reduce the overall energy footprint. Yet, carbon analysis of PID accelerators is covered for only a limited set of hardware and workloads. In this work, we present a carbon modeling framework for PID-enabled inference servers that captures embodied carbon along with idle and active operational carbon. Our evaluation spans three GPU baselines, six PID architectures, and eight transformer workloads. Additionally, we propose an aging-aware model that finds the carbon-optimal replacement period for PID accelerators over a 10-year horizon. Across evaluated configurations, our analysis shows that including PIDs in AI inference servers increases the total embodied carbon by at most 2.5\%, while the total carbon is reduced by up to 1.14$\times$ for encoder and 1.04$\times$ for decoder transformer workloads. The results also depend on GPU selection for pairing, where introducing PID modules in high-end HGX-H100 servers incurs 5.4--5.9$\times$ higher total carbon compared to A100 and H100 servers, due to their larger idle and embodied carbon. Introducing PID modules in higher-end GPU servers, compared to the A100-only baseline, increases (decreases) the total carbon-delay product (CDP) for encoder workloads (decoder workloads).

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.AR, nuovi lavori del giorno).
Categorie arXiv: Hardware Architecture (cs.AR) ; Emerging Technologies (cs.ET).
Pagina del lavoro: https://arxiv.org/abs/2610.27345 — 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 2026-10-21T07:06:28.105Z dalla pagina https://arxiv.org/list/cs.AR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).