Job scheduling techniques for distributed systems with temporal constraints

  • Authors:

  • Ping-Yi Lin;Pangfeng Liu

  • Affiliations:

  • Department of Computer Science and Information Engineering, National Taiwan University;,Department of Computer Science and Information Engineering, National Taiwan University

  • Venue:
  • GPC'10 Proceedings of the 5th international conference on Advances in Grid and Pervasive Computing
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Advance reservation of resources imporves quality-of-service by assuring the availability of resources at a future time However, it tends to fragment the available resources time and may lead to poor system utilization This paper proposes new job scheduling techniques to address this temporal fragmentation problem The goal is to schedule jobs with dependency to processors whose available times are fragmented into non-continuous time slots We propose a greedy algorithm to find the optimal schedule for jobs with single chain dependency We also propose a dynamic programming algorithm to find the optimal schedule for jobs with multiple-chain dependency In order to reduce scheduling time, we also propose three efficient heuristic algorithms Experimental results indicate that these heuristics are scalable and can find near optimal solutions.