A method of virtual machine placement for fault-tolerant cloud applications

摘要

Cloud applications are usually large scale and complicated. The placement of virtual machines (VMs) for highly reliable cloud applications is a challenging and critical research problem. To attack this challenge, a method of VM placement based on adaptive selection of fault-tolerant strategy for cloud applications is proposed. It involves two phase. In the first phase, the fault-tolerant strategies of cloud applications are sorted according to the constantly change of cloud applications constraint factors including the response time, failure rate and resource consumption. In the second phase, the VM placement plan based on adaptive selection of fault-tolerant strategy for cloud applications is solved. A prototype of VM placement framework based on adaptive selection of fault-tolerant strategy for cloud applications, named SelfAdaptionFTPlace, is implemented. Experimental results demonstrate that the proposed method shows up better performance and VM placement plan according to the constant change of cloud applications constraint factors compared with the existing methods.