RSIO: Automatic User Interaction Detection and Scheduling

Haoqiang Zheng, Jason Nieh

Proceedings of the ACM International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS 2010), New York, NY, June 14-18, 2010, pp. 263-274

Abstract

We present RSIO, a processor scheduling framework for im- proving the response time of latency-sensitive applications by monitoring accesses to I/O channels and inferring when user interactions occur. RSIO automatically identifies pro- cesses involved in a user interaction and boosts their prior- ities at the time the interaction occurs to improve system response time. RSIO also detects processes indirectly in- volved in processing an interaction, automatically account- ing for dependencies and boosting their priorities accord- ingly. RSIO works with existing schedulers and requires no application modifications to identify periods of latency- sensitive application activity. We have implemented RSIO in Linux and measured its effectiveness on microbenchmarks and real applications. Our results show that RSIO is easy to use and can provide substantial improvements in system performance for latency-sensitive applications.

PDF

sigmetrics2010:rsio

Columbia University Department of Computer Science