Paper Info
Title
A Study on Job Co-Allocation in Multiple HPC Clusters
Abstract
To more effectively use HPC clusters for even larger computations, improve turn-around times and better utilize compute resource, users are looking to interconnect multiple HPC clusters, creating a grid. To effectively use such grids, it may be desirable to split and co-allocate jobs requiring many processes across multiple clusters. While splitting a very large job across multiple clusters is an attractive possibility, the benefit, in terms of improving turn-around time, ultimately depends on the communication patterns between processes, workload on the communication links, and the maximum bandwidth of the links. The objective of this work is to understand the impact of communications on multi-processor jobs in order to develop scheduling strategies and job allocation algorithms for multi-cluster grids which can accommodate communication factors. In this paper we report on initial investigations of some co-allocation strategies. This evaluation is based on a simulator that has been implemented and validated experimentally across two HPC clusters.
Year
DOI
Venue
2006
10.1109/HPCS.2006.8
HPCS
Keywords
Field
DocType
job co-allocation,hpc cluster,co-allocate job,turn-around time,multiple hpc cluster,multiple hpc clusters,multiple cluster,communication link,communication factor,communication pattern,large job,job allocation algorithm,high performance computing,resource management,scheduling algorithm,grid computing,bandwidth,computer networks,utility computing,computer science,clustering algorithms
Resource management,Grid computing,Supercomputer,Computer science,Scheduling (computing),Workload,Parallel computing,Bandwidth (signal processing),Rate-monotonic scheduling,Grid,Distributed computing
Conference
ISSN
ISBN
Citations 
1550-5243
0-7695-2582-2
2
PageRank 
References 
Authors
0.39
8
2
Name
Order
Citations
PageRank
Jinhui Qin1143.82
Michael Bauer2729.60