Home // CLOUD COMPUTING 2014, The Fifth International Conference on Cloud Computing, GRIDs, and Virtualization // View article

Towards Makespan Minimization Task Allocation in Data Centers

Authors:
Kangkang Li
Ziqi Wan
Jie Wu
Adam Blaisse

Keywords: Task allocation; Data centers; Makespan

Abstract:
Nowadays, data centers suffer from resource limitations in both the limited bandwidth resources on the links and the computing capability on the servers, which triggers a variety of resource management problems. In this paper, we discuss one classic resource allocation problem: task allocation in data centers. That is, given a set of tasks with different makespans, how to schedule these tasks into the data center to minimize the average makespan. Due to the tradeoff between locality and load balancing, along with the multi-layer topology of data centers, it is extremely time consuming to obtain an optimal result. To deal with the multi-layer topology, we first study a simple case of one-layer cluster and discuss the optimal solution. After that, we propose our hierarchical task allocation algorithm for multi-layer clusters. Evaluation results prove the high efficiency of our algorithm.

Pages: 70 to 75

Copyright: Copyright (c) IARIA, 2014

Publication date: May 25, 2014

Published in: conference

ISSN: 2308-4294

ISBN: 978-1-61208-338-4

Location: Venice, Italy

Dates: from May 25, 2014 to May 29, 2014