A Study on Locality-Aware Request Distribution In Cluster-Based Network Servers

Improving Locality and Scalability in Cluster-Based Network Servers

Abstract

Weconsider cluster-based network servers in which a front-end directs inwardrequests to one of a number of back-ends. Specifically, we considercontent-based request distribution: the front-end uses the content requested,in addition to information about the load on the back-end nodes, to choosewhich back-end will handle this request. Content-based request distribution canimprove locality in the back-ends’ main memory caches, increase secondarystorage scalability by partitioning the server’s database, and provide theability to employ back-end nodes that are specialized for certain types ofrequests.