M
Morris A. Jette
Researcher at Lawrence Livermore National Laboratory
Publications - 10
Citations - 1692
Morris A. Jette is an academic researcher from Lawrence Livermore National Laboratory. The author has contributed to research in topics: Scheduling (computing) & Gang scheduling. The author has an hindex of 8, co-authored 10 publications receiving 1436 citations.
Papers
More filters
Book ChapterDOI
SLURM: Simple Linux Utility for Resource Management
TL;DR: A new cluster resource management system called Simple Linux Utility Resource Management (SLURM) is described in this paper, designed to be flexible and fault-tolerant and can be ported to other clusters of different size and architecture with minimal effort.
Book ChapterDOI
Improved Utilization and Responsiveness with Gang Scheduling
TL;DR: Empirical evidence from using gang scheduling on a Cray T3D installed at Lawrence Livermore National Lab corroborates these results, and shows conclusively that gang scheduling can be very effective with current technology.
Proceedings ArticleDOI
Performance Characteristics of Gang Scheduling in Multiprogrammed Environments
TL;DR: Gang scheduling provides both space-slicing and time slicing of parallel programs for better overall system responsiveness and utilization than otherwise possible as discussed by the authors, and the potential benefits of this technology to parallel processing are no less significant than time-sharing was in the 1960s.
Proceedings ArticleDOI
An Evaluation of Parallel Job Scheduling for ASCI Blue-Pacific
TL;DR: This paper analyzes the behavior of a gang-scheduling system that is developing for the ASCI Blue-Pacific machines and shows that both backfilling and gang- scheduling with moderate multiprogramming levels are much more effective than simple first-come first-serve scheduling.
Proceedings ArticleDOI
An Infrastructure for Efficient Parallel Job Execution in Terascale Computing Environments
TL;DR: The goal is to let the site scheduler control spatial allocation of jobs, if so desired, and to decide when jobs run, and the results show that higher multiprogramming levels lead to higher system utilization and lower job response times.