Assessing the Performance of the SRR Loop Scheduler with Irregular Workloads

The input workload of an irregular application must be evenly distributed amongits threads to enable cutting-edge performance. To address this need in OpenMP, several loopscheduling strategies were proposed. While having this ever-increasing number of strategies at dis-posal is helpful, it has become a non-trivial task to select the best one for a particular application.Nevertheless, this challenge becomes easier to be tackled when existing scheduling strategies areextensively evaluated. Therefore, in this paper, we present a performance and scalability eval-uation of the recently-proposed loop scheduling strategy named Smart Round-Robin (SRR). Todeliver a comprehensive analysis, we coupled a kernel benchmarking technique with several rigorousstatistical tools, and considered OpenMP’s Static and Dynamic loop schedulers as our baselines.Our results unveiled that SRR performs better on irregular applications with symmetric workloadsand coarse-grained parallelization, achieving up to 1.9x and 1.5x speedup over OpenMP’s Staticand Dynamic schedulers on synthetic kernels, respectively. On a N-Body Simulations applicationkernel, SRR delivered 2.48x better performance in contrast to OpenMP’s Dynamic scheduler