Modeling and Analysis of Dynamic Coscheduling in Parallel and Distributed Environments

Scheduling in large parallel systems continues to be an important and challenging research problem. Several key factors have resulted in the emergence of a new class of scheduling strategies: Dynamic coscheduling. Given their large design and performance spaces, it is difficult to fully explore the benefits and limitations of the proposed dynamic coscheduling approaches for large systems solely with the use of simulation or experimentation. We therefore formulate a general mathematical model of this class of scheduling strategies, and derive an exact and approximate matrix-analytic analysis for relatively small and large instances of the model, respectively. The results of numerical experiments with our approximation are in excellent agreement with detailed simulation results. Our analysis is then used to explore fundamental design and performance tradeoffs associated with the class of dynamic coscheduling policies across a broad spectrum of parallel computing environments.

By: Mark S. Squillante; Yanyong Zhang; Anand Sivasubramaniam; Natarajan Gautam; Hubertus Franke, Jose Moreira

Published in: RC23780 in 2005

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.

rc23780.pdf

Questions about this service can be mailed to reports@us.ibm.com .