Metraipolis › archive › dc_75509efafd63

Towards Efficient HPC Systems for Agents: Challenges and Opportunities — arXiv:2609.38723

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.38723
License (as archived)
CC0-1.0
Archived
2026-10-07T13:14:48.940Z
SHA-256 of the archived text
ac0cc6cb3fb31c1369aeb622be8f8cf0fd5e335972211dd464f8248fd8feb830
Stable URL of this page
https://metraipolis.com/documento/dc_75509efafd63

How to cite

Towards Efficient HPC Systems for Agents: Challenges and Opportunities — arXiv:2609.38723 — arXiv (en), https://arxiv.org/abs/2609.38723 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_75509efafd63 (Metraipolis archive dc_75509efafd63, sha256 ac0cc6cb3fb31c1369aeb622be8f8cf0fd5e335972211dd464f8248fd8feb830)
BibTeX
@misc{dc75509efafd63,
  title = {Towards Efficient HPC Systems for Agents: Challenges and Opportunities — arXiv:2609.38723},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38723},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_75509efafd63; sha256 ac0cc6cb3fb31c1369aeb622be8f8cf0fd5e335972211dd464f8248fd8feb830},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38723. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Coding agents have become real users of high-performance computing (HPC) systems, yet today's HPC abstractions, interfaces, and policies remain designed for human-driven workflows. In our measurement, users running coding agents are only 19.5% of the observed population, but account for 55.8% of job submissions, 29.1% of CPU core-hours, and 42.7% of GPU-hours. Agents are not simply faster humans. They issue commands at 20.8x the human rate, decompose work into fine-grained explore-modify-execute loops, and pursue open-ended goals through trial-and-error campaigns that continue through nights and weekends. These behaviors strain abstractions built for human timescales, surfacing as control-plane pressure on the scheduler, metadata-intensive I/O on bandwidth-provisioned filesystems, repeated rediscovery of what earlier sessions already learned, and new prompt-injection and policy-enforcement surfaces. Neither banning agents nor treating them as ordinary users is sustainable. We instead argue for co-design, that facilities should treat agents as first-class principals where agents become facility-aware tenants. We outline the resulting challenges and opportunities in compute, storage, agent memory, and safety.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-10-01 in cs.DC).
Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2609.38723 — 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 2026-10-07T12:46:39.709Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).