Metraipolis › archive › dc_7f2e23abc03a

Mosaic: GPU Sharing with Latency Guarantees through Kernel-Level Interference Prediction — arXiv:2610.07504

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07504
License (as archived)
CC0-1.0
Archived
2026-10-07T11:24:36.922Z
SHA-256 of the archived text
6f3b83e2f95c503bb47b3bf65c501bfce8549c52aacdbceb4505e506089d9d2f
Stable URL of this page
https://metraipolis.com/documento/dc_7f2e23abc03a

How to cite

Mosaic: GPU Sharing with Latency Guarantees through Kernel-Level Interference Prediction — arXiv:2610.07504 — arXiv (en), https://arxiv.org/abs/2610.07504 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7f2e23abc03a (Metraipolis archive dc_7f2e23abc03a, sha256 6f3b83e2f95c503bb47b3bf65c501bfce8549c52aacdbceb4505e506089d9d2f)
BibTeX
@misc{dc7f2e23abc03a,
  title = {Mosaic: GPU Sharing with Latency Guarantees through Kernel-Level Interference Prediction — arXiv:2610.07504},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07504},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7f2e23abc03a; sha256 6f3b83e2f95c503bb47b3bf65c501bfce8549c52aacdbceb4505e506089d9d2f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07504. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

GPUs are increasingly in demand for AI workloads, yet often remain substantially underutilized, motivating workload colocation. However, colocation introduces interference that can degrade latency-critical workloads. Existing approaches mitigate interference using either heuristic-based scheduling or interference predictors. Heuristics rely on coarse-grained metrics that overlook important interference sources, while predictors often depend on simulator-specific or similarly coarse metrics. However, GPU interference is complex and arises from multiple mechanisms, including thread-block placement, memory hierarchy contention, and intra-SM resource contention. We present Mosaic, a kernel-level interference predictor that explicitly models these mechanisms using a combination of analytical and lightweight learned models. Across four GPU architectures, Mosaic reduces prediction error by up to an order of magnitude compared to prior predictors. We integrate Mosaic into a scheduler, MosaicSched, that performs online kernel admission control and selects between full-GPU colocation and SM partitioning to maximize best-effort throughput while satisfying latency SLOs. Across all workloads, MosaicSched keeps the p99 latency below or very close to the target SLO.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.DC, nuovi lavori del giorno).
Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2610.07504 — 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 2026-10-07T11:15:54.469Z dalla pagina https://arxiv.org/list/cs.DC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).