Friendly virtual machines: leveraging a feedback-control model for application adaptation
Proceedings of the 1st ACM/USENIX international conference on Virtual execution environments
Adversarial exploits of end-systems adaptation dynamics
Journal of Parallel and Distributed Computing
Adaptive run-time performance optimization through scalable client request rate control
Proceedings of the 2nd ACM/SPEC International Conference on Performance engineering
Hi-index | 0.00 |
For QoS, we propose a new admission control mechanism called Q-PID, with which a web server system provides an user with satisfactory service by guaranteeing the bounded and predictive response time. The Q-PID controller is based on a feedback method that current output of the system is fed back into the input so that the system controls the next output properly. The controller accepts or rejects a new incoming request according to the accept ratio which is estimated dynamically at runtime by averaging the response times of previously accepted requests. Even though the system is under heavy load, the controller can sustain the bounded response time. We designed and implemented the controller for a cluster-based web server system. The experimental results indicate that the controller provides bounded response time.