Efficient distributed workstealing via matchmaking

Abstract

Many classes of high-performance applications and combinatorial problems exhibit large degree of runtime load variability. One approach to achieving balanced resource use is to over decompose the problem on fine-grained tasks that are then dynamically balanced using approaches such as workstealing. Existing work stealing techniques for such irregular applications, running on large clusters, exhibit high overheads due to potential untimely interruption of busy nodes, excessive communication messages and delays experienced by idle nodes in finding work due to repeated failed steals. We contend that the fundamental problem of distributed work-stealing is of rapidly bringing together work producers and consumers. In response, we develop an algorithm that performs timely, lightweight and highly efficient matchmaking between work producers and consumers which results in accurate load balance. Experimental evaluations show that our scheduler is able to outperform other distributed work stealing schedulers, and to achieve scale beyond what is possible with current approaches.

Toward Dynamic Load Balancing across OpenMP Thread Teams for Irregular Workloads

“…In addition to some centralized scheduling methods [18,30] for dynamic load balancing, work stealing [1,4,12,22] is a popular scheduling method. It is proved that work stealing is efficient for scheduling some multi-threaded executions.…”