On multi-user large-scale multi-computers, application workload is
highly variable and typically unpredicatable. In this paper, we present
and analyze the performance of three scheduling policies on a spectrum
of application workloads. Two of these policies make their scheduling
decisions at job start-up and the resources allocated to a job remain
constant until the job completion. The third policy exploits
application information in dynamically reconfiguring the resources
allocated to the job in order to achieve better system utilization and
application throughput. These policies are implemented in the context
of DRMS, a distributed resource management system that provides the
necessary infrastructure for dynamically varying application resources.
Our results show that the reconfigurable policy out-performs the former
two policies on a wide spectrum of workloads.
By: Ravi B. Konuru, Jose E. Moreira and Vijay K. Naik
Published in: RC20390 in 1996
LIMITED DISTRIBUTION NOTICE:
This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.
Questions about this service can be mailed to reports@us.ibm.com .