On caching effectiveness of web clusters under persistent connections
Journal of Parallel and Distributed Computing - Scalable web services and architecture
Future Generation Computer Systems - Special issue: Modeling and simulation in supercomputing and telecommunications
A web cluster simulator for performance analysis of the ALBM cluster system
AIS'04 Proceedings of the 13th international conference on AI, Simulation, and Planning in High Autonomy Systems
Performance comparison of various web cluster architectures
AsiaSim'04 Proceedings of the Third Asian simulation conference on Systems Modeling and Simulation: theory and applications
| Hi-index | 0.00 |
Abstract Server switches distribute incoming request traffic across the nodes of Internet server clusters and Web proxy cache arrays. These switches are a standard building block for large-scale Internet services, with many commercial products on the market. As Inter- net applications and service architectures continue to evolve, the role of server switches and the demands on their request r outing policies will also change. This paper explores interrelated factors shaping the role of server switching. These factors include the emergence of content delivery networks, the increasing prevalence of dynamic content, persistent connections in the HTTP 1.1 standard, the changing nature of Web server clusters, and the opportunities to apply server switching techniques to service protocols other than HTTP. We outline the potential role of server switches in virtualizing IP-based storage protocols and as a foundation for adaptive resource provisioning in server clusters.