Quantifying the cost of context switch
Abstract:
Measuring the indirect cost of context switch is a challenging problem. In this paper, we show our results of experimentally quantifying the indirect cost of context switch using a synthetic workload. Specifically, we measure the impact of program data size and access stride on context switch cost. We also demonstrate the potential impact of OS background interrupt handling on the measurement accuracy. Such impact can be  alleviated by using a multi-processor system on which one processor is employed for context switch measurement while the other runs OS background tasks.
This report will appear in the Workshop on Experimental Computer Science (ExpCS'07), Part of ACM FCRC, San Diego, CA, June 2007.
PDF version
Source code