IEEE Transactions on Computers
Web server performance measurement and modeling techniques
Performance Evaluation - Special issue on tools for performance evaluation
ACM SIGMETRICS Performance Evaluation Review
Analysis of SRPT scheduling: investigating unfairness
Proceedings of the 2001 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
IEEE Transactions on Software Engineering
A Multilayer Client-Server Queueing Network Model with Synchronous and Asynchronous Messages
IEEE Transactions on Software Engineering
SRPT Scheduling for Web Servers
JSSPP '01 Revised Papers from the 7th International Workshop on Job Scheduling Strategies for Parallel Processing
Size-based scheduling to improve web performance
ACM Transactions on Computer Systems (TOCS)
Web Performance Modeling Issues
International Journal of High Performance Computing Applications
Layered Bottlenecks and Their Mitigation
QEST '06 Proceedings of the 3rd international conference on the Quantitative Evaluation of Systems
Connection scheduling in web servers
USITS'99 Proceedings of the 2nd conference on USENIX Symposium on Internet Technologies and Systems - Volume 2
Computational Statistics & Data Analysis
A weighted-fair-queuing (WFQ)-based dynamic request scheduling approach in a multi-core system
Future Generation Computer Systems
Product-form results for two-station networks with shared resources
Performance Evaluation
Separation of timescales in a two-layered network
Proceedings of the 24th International Teletraffic Congress
Modeling server usage for online ticket sales
Proceedings of the Winter Simulation Conference
Hi-index | 0.00 |
Popular web sites are expected to handle huge number of requests concurrently within a reasonable time frame. The performance of these web sites is largely dependent on effective thread management of their web servers. Although the implementation of static and dynamic thread policies is common practice, remarkably little is known about the implications on performance. Moreover, the commonly used policies do not take into account the complex interaction between the threads that compete for access to a shared resource. We propose new dynamic thread-assignment policies that minimize the average response time of web servers. The web server is modeled as a two-layered tandem of multi-threading queues, where the active threads compete for access to a common resource. This type of two-layered queueing model, which occurs naturally in the performance modeling of systems with intensive software-hardware interaction, are on the one hand appealing from an application point of view, but on the other hand are challenging from a methodological point of view. Our results show that the optimal dynamic thread-assignment policies yield strong reductions in the response times. Validation on an Apache web server shows that our dynamic thread policies confirm our analytical results.