Session <Full Schedule · Contributors · Organizations · Search ProgramPaper: Resource ManagementSession ChairAnn GentileSandia National LaboratoriesEvent TypePaperTimeTuesday, 14 November 20231:30pm - 3pm Location401-402Tags Architecture and Networks Performance Measurement, Modeling, and Tools Resource Management Registration Categories TP Reproducibility BadgesPresentations1:30pm - 2:00pm Mirage: Toward Low-interruption Services on Batch GPU Clusters with Reinforcement LearningAuthorsQiyang DingPengfei ZhengShreyas KudariShivaram VenkataramanZhao Zhang2:00pm - 2:30pm Prodigy: Toward Unsupervised Anomaly Detection in Production HPC SystemsAuthorsBurak AksarEfe SencanBenjamin SchwallerOmar AazizVitus J. LeungJim BrandtBrian KulisManuel EgeleAyse K. Coskun2:30pm - 3:00pm DPS: Adaptive Power Management for Overprovisioned SystemsAuthorsJianru DingHenry Hoffmann
