Shortfall-Aware Decomposed Allocation with Repair for Heterogeneous GPU Clusters: Reproduction Artifact

Fuente: Zenodo
Salvato in:
Dettagli Bibliografici
Autore principale: Dorn, Yuriy
Natura: Recurso digital
Lingua:inglese
Pubblicazione: Zenodo 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866901729590640640
author Dorn, Yuriy
author_facet Dorn, Yuriy
contents <p>This record contains the reproduction artifact for the manuscript “Shortfall-Aware Decomposed Allocation with Repair for Heterogeneous GPU Clusters.”</p> <p>The archive includes:<br>- manuscript source and PDF<br>- runnable simulator code and evaluation scripts<br>- final paper figures<br>- cached JSON/CSV outputs used for the reported experiments<br>- bounded direct replay summaries for real Philly replay windows</p> <p>The artifact supports the paper’s exact small-instance benchmark, dynamic scaling study, fairness analysis, ablations, and bounded direct replay checks.</p> <p>Raw Alibaba and Philly trace files are not redistributed in this record. Only derived aggregate replay outputs are included. Raw traces must be obtained separately from their original sources and are not part of this upload.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_19544795
institution Zenodo
language eng
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle Shortfall-Aware Decomposed Allocation with Repair for Heterogeneous GPU Clusters: Reproduction Artifact
Dorn, Yuriy
GPU clusters
Resource Allocation
scheduling
fairness
ADMM
<p>This record contains the reproduction artifact for the manuscript “Shortfall-Aware Decomposed Allocation with Repair for Heterogeneous GPU Clusters.”</p> <p>The archive includes:<br>- manuscript source and PDF<br>- runnable simulator code and evaluation scripts<br>- final paper figures<br>- cached JSON/CSV outputs used for the reported experiments<br>- bounded direct replay summaries for real Philly replay windows</p> <p>The artifact supports the paper’s exact small-instance benchmark, dynamic scaling study, fairness analysis, ablations, and bounded direct replay checks.</p> <p>Raw Alibaba and Philly trace files are not redistributed in this record. Only derived aggregate replay outputs are included. Raw traces must be obtained separately from their original sources and are not part of this upload.</p>
title Shortfall-Aware Decomposed Allocation with Repair for Heterogeneous GPU Clusters: Reproduction Artifact
topic GPU clusters
Resource Allocation
scheduling
fairness
ADMM
url https://doi.org/10.5281/zenodo.19544795