Decentralized Load Balancing for Heterogeneous Grids

Author

Al-Azzoni, Issam ; Down, Douglas G.

Author_Institution

Dept. of Comput. & Software, McMaster Univ., Hamilton, ON, Canada

fYear

2009

fDate

15-20 Nov. 2009

Firstpage

545

Lastpage

550

Abstract

Several decentralized load balancing policies have been proposed to address the issue of scalability in grids.However, the communication overhead incurred in exchanging state information remains a burden. In this paper, we propose a dynamic, decentralized load balancing policy which performs very competitively in heterogeneous grids. The policy uses an effective mechanism for state information exchange, which significantly reduces the communication overhead, while quickly updating the state information in a decentralized fashion.

Keywords

grid computing; resource allocation; decentralized load balancing policies; distributed systems; grid computing; heterogeneous grids; state information exchange; Availability; Computer networks; Distributed computing; Grid computing; Hardware; High performance computing; IP networks; Linear programming; Load management; Scalability; computing; decentralized load balancing; distributed systems; heterogeneous processors; information exchange;

fLanguage

English

Publisher

ieee

Conference_Titel

Future Computing, Service Computation, Cognitive, Adaptive, Content, Patterns, 2009. COMPUTATIONWORLD '09. Computation World:

Conference_Location

Athens

Print_ISBN

978-1-4244-5166-1

Electronic_ISBN

978-0-7695-3862-4

Type

conf

DOI

10.1109/ComputationWorld.2009.35

Filename

5359648