ScienceDirect® Home Skip Main Navigation Links
You have guest access to ScienceDirect. Find out more.
 
Home
Browse
My Settings
Alerts
Help
 Quick Search
 Search tips (Opens new window)
    Clear all fields    
advertisementadvertisement
Future Generation Computer Systems
Volume 21, Issue 1, 1 January 2005, Pages 135-149
 
Font Size: Decrease Font Size  Increase Font Size
 Abstract - selected
Article
Purchase PDF (415 K)

 
 
 
Related Articles in ScienceDirect
View More Related Articles
 
View Record in Scopus
 
doi:10.1016/j.future.2004.09.032    How to Cite or Link Using DOI (Opens New Window)
Copyright © 2004 Elsevier B.V. All rights reserved.

Grid load balancing using intelligent agents

Junwei Caoa, 1, Corresponding Author Contact Information, E-mail The Corresponding Author, Daniel P. Spoonerb, Stephen A. Jarvisb and Graham R. Nuddb

aCenter for Space Research, Massachusetts Institute of Technology, Cambridge, MA, USA bDepartment of Computer Science, University of Warwick, Coventry, UK

Available online 28 October 2004.

Purchase the full-text article



References and further reading may be available for this article. To view references and further reading you must purchase this article.

Abstract

Scalable management and scheduling of dynamic grid resources requires new technologies to build the next generation intelligent grid environments. This work demonstrates that AI techniques can be utilised to achieve effective workload and resource management. A combination of intelligent agents and multi-agent approaches is applied to both local grid resource scheduling and global grid load balancing. Each agent is a representative of a local grid resource and utilises predictive application performance data with iterative heuristic algorithms to engineer local load balancing across multiple hosts. At a higher level, agents cooperate with each other to balance workload using a peer-to-peer service advertisement and discovery mechanism.

Keywords: Load balancing; Grid computing; Intelligent agents; Genetic algorithm; Service discovery

Article Outline

1. Introduction
2. Grid agents
2.1. Agent structure
2.2. Agent hierarchy
2.3. Performance prediction
3. Local grid load balancing
3.1. First-come-first-served algorithm
3.2. Genetic algorithm
4. Global grid load balancing
4.1. Service advertisement and discovery
4.2. System implementation
5. Performance evaluation
5.1. Performance metrics
5.1.1. Total application execution time
5.1.2. Average advance time of application execution completion
5.1.3. Average resource utilisation rate
5.1.4. Load-balancing level
5.2. Experimental design
5.3. Experimental results
5.3.1. Experiment 1
5.3.2. Experiment 2
5.3.3. Experiment 3
5.3.4. Application execution
5.3.5. Resource utilisation
5.3.6. Load balancing
5.4. Agent performance
5.4.1. Total application execution time
5.4.2. Average advance time of application execution completion
5.4.3. Network packets
6. Related work
7. Conclusions
References
Vitae












 
Home
Browse
My Settings
Alerts
Help
Elsevier.com (Opens new window)
About ScienceDirect  |  Contact Us  |  Information for Advertisers  |  Terms & Conditions  |  Privacy Policy
Copyright © 2008 Elsevier B.V. All rights reserved. ScienceDirect® is a registered trademark of Elsevier B.V.