Distributed computing feasibility in a non-dedicated homogeneous distributed system

Abstract

The authors address the feasibility of a nondedicated parallel processing environment, assuming workstation processes have preemptive priority over parallel tasks. They develop an analytical model to predict parallel job response times. The model provides insight into how significantly the workstation owner interference degrades parallel program performance. A new term, task ratio, which relates the parallel task demand to the mean service demand of nonparallel workstation processes, is introduced. It is proposed that the task ratio is a useful metric for determining how large the demand of a parallel applications must be in order to make efficient use of a nondedicated distributed system.