Metraipolis › archive › dc_f2c99a1d7992

COMPASS-ABS: Reducing Fragmentation in Shared GPU Clusters for Deep Learning Training Workloads — arXiv:2609.18519

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18519
License (as archived)
CC0-1.0
Archived
2026-10-15T13:53:39.377Z
SHA-256 of the archived text
628390481cde0e90f7958c94774818ddcf931a6ea23d71c3df7633d5e04fa550
Stable URL of this page
https://metraipolis.com/documento/dc_f2c99a1d7992

How to cite

COMPASS-ABS: Reducing Fragmentation in Shared GPU Clusters for Deep Learning Training Workloads — arXiv:2609.18519 — arXiv (en), https://arxiv.org/abs/2609.18519 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f2c99a1d7992 (Metraipolis archive dc_f2c99a1d7992, sha256 628390481cde0e90f7958c94774818ddcf931a6ea23d71c3df7633d5e04fa550)
BibTeX
@misc{dcf2c99a1d7992,
  title = {COMPASS-ABS: Reducing Fragmentation in Shared GPU Clusters for Deep Learning Training Workloads — arXiv:2609.18519},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18519},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f2c99a1d7992; sha256 628390481cde0e90f7958c94774818ddcf931a6ea23d71c3df7633d5e04fa550},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18519. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

With the rapid advancement of deep learning technology, shared GPU clusters receive an increasing number of deep learning training (DLT) jobs. Yet resource fragmentation make such clusters underutilized and forces the DLT jobs running on them to endure long turnaround times. Extensive research has been devoted to quantifying fragmentation and developing scheduling algorithms that alleviate its impact. However, existing fragmentation measures break down in the absence of workload distribution information, while current schedulers cannot continuously maintain resource fragmentation at a low level. To tackle these problems, we first introduce Scheduler-Induced Fragmentation (SIF), a metric built on the notion of partial-nodes that is independent of historical workload knowledge. We then propose COMPASS-ABS, which employs the COMPact-ASSured (COMPASS) algorithm to confine the cluster state within a tight Anchor-Based Space (ABS), whose construction fully leverages the topological alignment between dominant workload size and node capacity. Moreover. We also prove that it ensures SIF is bounded by $\frac{2}{N}$ under a workload composition condition that matches both theory and production. Evaluations implemented on a physical cluster and a simulated cluster demonstrate COMPASS-ABS effectiveness at improving resource utilization, reducing DLT job completion time by reducing fragmentation.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-17 in cs.DC).
Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.18519 — 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 2026-10-15T13:48:52.625Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).