Web content adaptation to improve server overload behavior
WWW '99 Proceedings of the eighth international conference on World Wide Web
Performance Guarantees for Web Server End-Systems: A Control-Theoretical Approach
IEEE Transactions on Parallel and Distributed Systems
Lessons from Giant-Scale Services
IEEE Internet Computing
Managing Web server performance with AutoTune agents
IBM Systems Journal
Analysis of Caching and Replication Strategies for Web Applications
IEEE Internet Computing
Personalizable edge services for Web accessibility
Universal Access in the Information Society
Communications of the ACM - Surviving the data deluge
Empirical quantification of opportunities for content adaptation in web servers
Proceedings of the 3rd Annual Haifa Experimental Systems Conference
MIMOSA: context-aware adaptation for ubiquitous web access
Personal and Ubiquitous Computing
Hi-index | 0.00 |
A basic problem in managing web servers is capacity planning. A partial solution is to use content adaptation, where the system automatically trades off quality for throughput, e.g. by eliminating graphical decorations and adjusting page layout. We evaluate this approach based on a full implementation in Apache and increasing load patterns. The implementation uses two alternative versions of the files, and employs URL rewriting rules to select which version to use. Triggering a switch from one version to the other is done based on readily available load metrics. The experiments show that throughput can be increased by a factor of 2 to 4 at the price of minor to acceptable deterioration in graphical quality. Increasing throughput by an order of magnitude is also possible, but requires larger compromises. Nevertheless, this is still achievable without a real effect on content. Thus content adaptation is a viable tool, but may be insufficient by itself for handling huge surges in load such as flash crowds.