Metraipolis › archive › dc_e99fe5afdf96
Provisioning Networks for AI Supercomputers: A Trace-Driven Study of Performance Sensitivity at Unprecedented Scale — arXiv:2610.41700
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.41700
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-02T07:55:04.761Z
- SHA-256 of the archived text
dbfbf60fd2fc01216d190962b1482274354ef6b6fc94799b9983ff153bc23df9- Stable URL of this page
https://metraipolis.com/documento/dc_e99fe5afdf96
How to cite
Provisioning Networks for AI Supercomputers: A Trace-Driven Study of Performance Sensitivity at Unprecedented Scale — arXiv:2610.41700 — arXiv (en), https://arxiv.org/abs/2610.41700 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e99fe5afdf96 (Metraipolis archive dc_e99fe5afdf96, sha256 dbfbf60fd2fc01216d190962b1482274354ef6b6fc94799b9983ff153bc23df9)
BibTeX
@misc{dce99fe5afdf96,
title = {Provisioning Networks for AI Supercomputers: A Trace-Driven Study of Performance Sensitivity at Unprecedented Scale — arXiv:2610.41700},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.41700},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e99fe5afdf96; sha256 dbfbf60fd2fc01216d190962b1482274354ef6b6fc94799b9983ff153bc23df9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41700. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
AI workloads on supercomputers make interconnect performance a primary design constraint. Hardware specifications and microbenchmarks alone do not show how real workloads respond to latency, bandwidth, and jitter. We present a trace-driven study of network sensitivity for training and inference workloads at scales up to 4,096 GPUs. Our methodology combines signature-based trace decomposition with linear programming to make broad network-parameter sweeps feasible. It preserves the recorded NCCL schedules and program dependencies, and estimates a lower bound on runtime under the LogGP model. We validate the compositional analysis against monolithic linear programs, LogGOPSim, and hardware measurements at feasible scales. The resulting latency-bandwidth envelopes identify low-sensitivity regions and performance cliffs that vary substantially across workloads. They also show when bandwidth upgrades provide meaningful returns and when additional network performance has little effect. These results translate execution traces into practical guidance for evaluating network configurations while making the model assumptions explicit. Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.NI, nuovi lavori del giorno). Categorie arXiv: Networking and Internet Architecture (cs.NI). Pagina del lavoro: https://arxiv.org/abs/2610.41700 — 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 2026-11-02T07:40:30.035Z dalla pagina https://arxiv.org/list/cs.NI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).