Case Study on Co-scheduling for HPC Applications

Most applications running on supercomputers achieve only a fraction of the peak performance of the system. In this paper we analyze the performance and energy efficiency of co-scheduling one memory bandwidth bound and one compute bound application on the same node. We present auto pin+, a tool designed to monitor and optimize co-scheduling of applications. Our analysis shows that co-scheduling can improve both energy efficiency and overall throughput of a supercomputer. At best, runtime can be decreased by 28% and the energy consumption by 12%, respectively, compared to best case dedicated execution. The overall efficiency however strongly depends on the ratio of jobs available in the queue. We furthermore present a simple adaptive strategy depending on the available jobs in the queue.