Towards Informed Web Content Delivery

Abstract. A wide range of techniques have been proposed, implemented, and even standardized for improving the performance of Web content delivery. However, previous work has found that many Web sites either do not take advantage of such techniques or unknowingly inhibit their use. In this paper, we present the design of a tool called Cassandra that addresses these problems. Web site developers can use Cassandra to achieve three goals: (i) to identify protocol correctness and conformance problems; (ii) to identify content delivery performance problems; and (iii) to evaluate the potential benefits of using content delivery optimizations. Cassandra combines performance and behavioral data, together with an extensible simulation architecture, to identify content delivery problems and predict optimization benefits. We describe the architecture of Cassandra and demonstrate its use to evaluate the potential benefits of a CDN on a large Web server farm.

• Type: Conference
• Year: 2004
• Where: WCW
• Authors: Leeann Bent, Michael Rabinovich, Geoffrey M. Voelker, Zhen Xiao
• Keywords: Cassandra, Content Delivery, Internet Technology, WCW 2004, Web Content Delivery