heSRPT

optimal scheduling of parallel jobs with known sizes

Benjamin Berg, Rein Vesilo, Mor Harchol-Balter

Research output: Contribution to journalConference paper

Abstract

Nearly all modern data centers serve workloads which are capable of exploiting parallelism. When a job parallelizes across multiple servers it will complete more quickly, but jobs receive diminishing returns from being allocated additional servers. Because allocating multiple servers to a single job is inefficient, it is unclear how best to share a fixed number of servers between many parallelizable jobs. In this paper, we provide the first closed form expression for the optimal allocation of servers to jobs. Specifically, we specify the number of servers that should be allocated to each job at every moment in time. Our solution is a combination of favoring small jobs (as in SRPT scheduling) while still ensuring high system efficiency. We call our scheduling policy high-efficiency SRPT (heSRPT).

Original languageEnglish
Pages (from-to)18-20
Number of pages3
JournalPerformance Evaluation Review
Volume47
Issue number2
DOIs
Publication statusPublished - Sep 2019
EventWorkshop on MAthematical performance Modeling and Analysis (21st : 2019) - Phoenix, United States
Duration: 28 Jun 201928 Jun 2019

Fingerprint Dive into the research topics of 'heSRPT: optimal scheduling of parallel jobs with known sizes'. Together they form a unique fingerprint.

Cite this