A Survey and Evaluation of Fault Tolerant Techniques of Virtual Machine in Cloud Computing Environment

With the rapid development of cloud computing, virtualization technology has experienced considerable progress and application. Virtualization technology allows not only an application, but the operating system and the hardware as well, to be encapsulated into a virtual machine (VM). Although virtualization technology has been widely adopted in cloud computing, there are still many research issues to be fully addressed. Frequent occurrence of VM outage incidents indicates that fault tolerant techniques of VM are needed to achieve high availability of cloud computing infrastructure. This paper introduces techniques related to fault tolerance of VM, surveys existing fault tolerant techniques based on live VM migration. Then it evaluates some of these techniques under kernel-build workload in terms of three performance evaluation metrics: downtime, total migration time (TMT), total data transmitted (TDT). Finally, this paper looks into future researches in this field. This comparative study will help VM researchers to better understand the behavior and performance of different fault tolerant techniques of VM in detail.