Dynamic coscheduling on workstation clusters

Abstract

Coscheduling has been shown to be a critical factor in achieving efficient parallel execution in timeshared environments [12, 19, 4]. However, the most common approach, gang scheduling, has limitations in scaling, can compromise good interactive response, and requires that communicating processes be identified in advance. We explore a technique called dynamic coscheduling (DCS) which produces emergent coscheduling of the processes constituting a parallel job. Experiments are performed in a workstation environment with high performance networks and autonomous timesharing schedulers for each CPU. The results demonstrate that DCS can achieve effective, robust coscheduling for a range of workloads and background loads. Empirical comparisons to implicit scheduling and uncoordinated scheduling are presented. Under spin-block synchronization, DCS reduces job response times by up to 20% over implicit scheduling while maintaining fairness; and under spinning synchronization, DCS reduces job response times by up to two decimal orders of magnitude over uncoordinated scheduling. The results suggest that DCS is a promising avenue for achieving coordinated parallel scheduling in an environment that coexists with autonomous node schedulers.