A Response Time based Load Balancing Algorithm for Service Composition

Author

Yan, CaiRong ; Zhu, Ming ; Shi, Youqun

Author_Institution

Sch. of Comput. Sci. & Technol., Donghua Univ., Shanghai

Volume

1

fYear

2008

fDate

6-8 Oct. 2008

Firstpage

13

Lastpage

16

Abstract

How to select candidate services and compose them is a key problem in Web service composition. In this paper, a response time based load balancing (RTLB) algorithm for service composition is proposed. Aiming at keeping the load balancing of node so as to achieve the largest throughput of the system, RTLB algorithm adopts the service load and node load as the decision-making measures and forwards requests to suitable nodes. Node load is made up of the services being disposing by the node. Service load, a ratio, can be obtained by self-learning mechanism in the course of comparing the response time among services continually. As the services deployed on the network are alterable, the service load will be adjusted dynamically. The experiment shows that by thinking the service load into account, RTLB algorithm can achieve better load balancing performance and quicker response time.

Keywords

Web services; decision making; resource allocation; unsupervised learning; RTLB algorithm; Web service composition; decision-making measure; response time based load balancing algorithm; self-learning mechanism; Algorithm design and analysis; Computer science; Delay; Filtering; Load management; Quality of service; Scheduling algorithm; Throughput; Utility theory; Web services; Load Balancing; Response Time; Self-learning; Service Composition;

fLanguage

English

Publisher

ieee

Conference_Titel

Pervasive Computing and Applications, 2008. ICPCA 2008. Third International Conference on

Conference_Location

Alexandria

Print_ISBN

978-1-4244-2020-9

Electronic_ISBN

978-1-4244-2021-6

Type

conf

DOI

10.1109/ICPCA.2008.4783561

Filename

4783561