Parallel Processing-Oriented Hybrid Scheduling of Virtual Machines in Cloud

Abstract

ABSTRACT In cloud computing environment, parallel applications generally run on symmetric multiprocessing (SMP) virtual machine (VM). Since this type of application requires synchronous operations between processes/threads, all virtual CPUs (vCPUs) of a parallel VM (i.e., the VM running parallel application) should be online simultaneously. At present, relevant studies have been intensively conducted from the perspective of vCPU co-scheduling in virtual machine monitor (VMM). However, the existing co-scheduling methods have the problems of unrestricted preemptions between parallel VMs, which probably results in negative impact on the performance of parallel applications in these VMs. To address the above problems, in this paper, we first analyze the deficiencies of the existing co-scheduling approaches in virtualized environment. Then we propose an enhanced co-scheduling algorithm to improve the performance of parallel application in SMP VM.