AASTMT

Content on this page requires a newer version of Adobe Flash Player.

Get Adobe Flash player

 
 

Distributed Systems

  • M. Kholief, I Awad, M. El-Derini, M. Nagi, H. Ali. Behavior of Task Scheduling Algorithms for Probabilistic Module Execution Times, ISCC’98, Athens, Greece, June 1998.

Abstract: Static task scheduling in distributed computing systems is a very complex problem and known to be NP-hard. This problem is even harder when the module execution times become probabilistic. In this paper we study the effect of probabilistic module execution times on the performance of task-scheduling algorithms. We show that in static task scheduling, for probabilistic module execution times, and in the existence of some factors there is no need to use an expensive task-scheduling algorithm. Given any two static task-scheduling algorithms that use deterministic module execution times in assigning task modules to the distributed system, the performance of these two algorithms will not remain the same when these module execution times become probabilistic rather than deterministic. We also study the effects of some factors an our results