In this paper we present a study of the job arrival patterns from a paralle
l computing system and the impact of such arrival patterns on the performan
ce of parallel scheduling strategies. Using workload data from the Cornell
Theory Center, we develop a class of traffic models to characterize these a
rrival patterns. Our analysis of the job arrival data illustrates traffic p
atterns that exhibit heavy-tailed behavior and other characteristics which
are quite different from the arrival processes used in previous studies of
parallel scheduling. We then investigate the impact of these arrival traffi
c patterns on the performance of parallel space-sharing strategies, includi
ng the derivation of some scheduling optimality results. (C) 1999 Published
by Elsevier Science B.V. All rights reserved.