Performance Comparison of Assignment Policies on Cluster-based E-Commerce Servers

Abstract: Cluster-based server architectures combine good performance and low cost, and are commonly used for applications that generate heavy loads. Essentially, a cluster-based server consists of a front-end dispatcher and several back-end servers. The dispatcher receives incoming requests, and then assigns them to back-end servers, which are responsible for request processing. The many benets of cluster-based servers make them a good choice for e-commerce applications as well. However, applying this type of architecture to e-commerce applications is hindered by the fact that e-commerce clusters have the additional task of verifying that requests comply with contract terms. The problem is further complicated by the fact that contract terms may be expressed as functions of dynamic, mutable states. We have proposed a policy, called TDA (Type Dependent Assignment), which is designed to balance load among back-end servers and to that request validation is done in a an ecient manner. In this paper we are interested in evaluating the performance of TDA. To gauge its ecacy, we present an empirical study, using real-life trace data measured at Internet cluster-based servers. Experimental results of this study show that TDA outperforms the Least Connected policy by a factor of four