Research on rejuvenation analytical models for a virtualized system with live VM migration

Yi Zhong, Jian Xu, Jing Zhong, Fengyu Liu

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12A) 439-445

School of Computer Science and Engineering Nanjing University of Science and Technology, Nanjing, Jiangsu, China, 210094

With the widespread use of virtualization technology, availability of Virtual Machines (VMs) in server virtualized systems became an important issue. In aspect of availability of virtualized systems, software rejuvenation is a favourable technology, which can postpone or prevent failures caused by software aging in VMs and underlying Virtual Machine Monitor (VMM). During the rejuvenation of the VMM, live VM migration can further improve availability of virtualized systems. In this paper, we will analysis rejuvenation process of virtualized system with live VM migration, propose a time and load based rejuvenation analytical model in Stochastic Reward Net (SRN) to describe state change during system rejuvenation, find the optimum combinations of rejuvenation trigger intervals that maximize the availability of VM, and do a experience to analyze and compare the model in this paper and the time based model. The experience result shows the time and load based rejuvenation analytical model is better than the time based model in respect of system availability and throughput rate, and is more stable in face of the dynamic change of system load.