Reducing the Impact of Interjob Interference in Dragonfly Networks Using Virtual Partitions

In high-performance computing systems, the interference among applications that share network resources is one of the major causes of performance degradation and increase variability among different executions of the same application. In this article, we propose a virtual partitioning scheme for dragonfly networks that combines a job-allocation policy with a virtual channel (VC) assignment. Our experiments confirm that our proposal offers a better performance than the established schemes, also reducing variability.