A Bee Colony based optimization approach for simultaneous job scheduling and data replication in grid environments

Academic Article

Abstract

  • This paper presents a novel Bee Colony based optimization algorithm, named Job Data Scheduling using Bee Colony (JDS-BC). JDS-BC consists of two collaborating mechanisms to efficiently schedule jobs onto computational nodes and replicate datafiles on storage nodes in a system so that the two independent, and in many cases conflicting, objectives (i.e., makespan and total datafile transfer time) of such heterogeneous systems are concurrently minimized. Three benchmarks - varying from small- to large-sized instances - are used to test the performance of JDS-BC. Results are compared against other algorithms to show JDS-BC's superiority under different operating scenarios. These results also provide invaluable insights into data-centric job scheduling for grid environments. © 2011 Elsevier Ltd.
  • Authors

    Published In

    Digital Object Identifier (doi)

    Author List

  • Taheri J; Choon Lee Y; Zomaya AY; Siegel HJ
  • Start Page

  • 1564
  • End Page

  • 1578
  • Volume

  • 40
  • Issue

  • 6