Scheduling Optimization for Resource-Intensive Web Requests on Server Clusters

Clustering supportwith a single-systemimagefor large-scaleWeb serversis important to improve the systemscalability in processing alargenumberof concurrentrequestsfrom Internet,especially when those requestsinvolve resource-intensivedynamic content generation.This paperproposesschedulingoptimization for aWeb serverclusterwith amaster/slavearchitecturewhich separatesstatic and dynamic content processing. Our experimental results show that the proposedoptimization using reservation-basedscheduling canproduceup to a68%performanceimprovement.

• Type: Conference
• Year: 1999
• Where: SPAA
• Authors: Huican Zhu, Ben Smith, Tao Yang
• Keywords: Distributed And Parallel Computing, Dynamic Content Processing, Requestsinvolve Resource-intensivedynamic Content, Single-systemimagefor Large-scaleWeb Serversis, SPAA 1999