Title :
Improvement of high reliable cluster load balancing algorithm based on multiple regression
Author :
Qiu, Zhao ; Chen, Mingrui ; Huang, Jun
Author_Institution :
Coll. of Econ. & Bus. Adm., Univ. of Chongqing, Chongqing
Abstract :
By the popularization of the Internet application, the performance requirement of Web servers is becoming higher and higher. Using several hosts as the cluster to provide services together is the common scheme for the moment, with high performance-price ratio, high reliability and scalability, while the key performance of the cluster system is the balancing algorithm. Based on the LVS project, this paper uses the multiple-linear regression to figure out the loading capacity of each host in the cluster and then balance the loading distribution to increase the loading capacity of the whole Web cluster.
Keywords :
Internet; file servers; pattern clustering; regression analysis; resource allocation; Internet application; Web cluster; Web servers; balancing algorithm; cluster load balancing algorithm; multiple regression; multiple-linear regression; performance-price ratio; service providers; Cluster; Load Balance; Multiple Regression;
Conference_Titel :
Service Operations and Logistics, and Informatics, 2008. IEEE/SOLI 2008. IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-2012-4
Electronic_ISBN :
978-1-4244-2013-1
DOI :
10.1109/SOLI.2008.4683045