Load Balancing in Arbitrary Network Topologies with Stochastic Adversarial Input

Abstract

We study the long-term (steady state) performance of a simple, randomized, local load balancing technique under a broad range of input conditions. We assume a system of n processors connected by an arbitrary network topology. Jobs are placed in the processors by a deterministic or randomized adversary. The adversary knows the current and past load… (More)
DOI: 10.1137/S0097539703437831

Topics

  • Presentations referencing similar topics