Study of an iterative resource allocation technique to minimize machine completion times in a distributed computing system
Date
2006
Authors
Oltikar, Mohana, author
BriceƱo, Luis D., author
Maciejewski, Anthony A., author
Siegel, Howard Jay, author
Journal Title
Journal ISSN
Volume Title
Abstract
Heterogeneous computing (HC) is the coordinated use of different types of machines, networks, and interfaces to maximize the combined performance and/or cost effectiveness. Heuristics for allocating resources in an HC system have different optimization criteria. A common optimization criterion is to minimize the completion time of the last to finish machine (makespan). In some environments, it is useful to minimize the finishing times of the other machines in the system, i.e., those machines that are not the last to finish. Consider a production environment where a set of known tasks are to be mapped to resources off-line before execution begins. In this study, we present an "iterative" approach for decreasing the finishing time of each machine in a given resource allocation, by repeatedly running a mapping heuristic to minimize makespan on all machines and then the non-makespan machines; i.e., ignoring the current makespan machine and the tasks assigned to it. This work identifies heuristics that can offer improvements in the completion time of non-makespan machines using this "iterative" approach.
Description
Rights Access
Subject
computing
allocation
heterogeneous
iterative