In collective I/O, MPI processes exchange requests so that the rearranged requests can result in the shortest file system access time. Scheduling the exchange sequence determines the response time of participating processes. Existing implementations that simply follow the increasing order of file ofsets do not necessary produce the best performance. To minimize the average response time, we propose three scheduling algorithms that consider the number of processes per file stripe and the number of accesses per process. Our experimental results demonstrate improvements of up to 50% in the average response time using two synthetic benchmarks and a high-resolution climate application.
Revised: December 14, 2012 |
Published: September 21, 2011
Citation
Jin C., S. Sehrish, W. Liao, A. Choudhary, and K.L. Schuchardt. 2011.Improving the Average Response Time in Collective I/O. In Recent Advances in the Message Passing Interface - Proceedings of the 18th European MPI Users’ Group Meeting (EuroMPI 2011), September 18-21, 2011, Santorini, Greece. Lecture Notes in Computer Science, edited by Y Cotronis, et al, 6960, 71-80. Heidelberg:Springer.PNNL-SA-80736.doi:10.1007/978-3-642-24449-0_10