Engineering server-driven consistency for large scale dynamic Web services

Recent research has shown that server-driven protocols for achieving cache consistency in wide-area network services can perform significantly better than traditional consistency protocols based on HTTP client polling. In this paper, we study how to engineer server-driven consistency solutions for large scale dynamic web services. The workload used in this study comes from IBM's Olympics web server, one of the most popular web servers on the Internet. Our study focuses on scalability and cachability of dynamic content. To assess scalability, we measure both the amount of state that a server needs to maintain to ensure consistency, and the bursts of load that a server sustains to send out invalidation messages whenever a popular object is modified. We find that it is possible to limit the size of the server's state without significant performance costs, and that bursts of load can be smoothed out with minimal impact on the consistency guarantees. To improve performance, we sy...

• Type: Conference
• Year: 2001
• Where: WWW
• Authors: Jian Yin, Lorenzo Alvisi, Michael Dahlin, Arun Iyengar
• Keywords: Cache Consistency, Internet Technology, Server-driven Consistency Protocols, Traditional Consistency Protocols, WWW 2001