Exploring the effects of virtual machine placement on the transmission of infections in cloud

Cloud Computing has recently emerged as a computing paradigm for delivering IT resources as services over the internet. However, there still exist significant problems with regard to secure delivery of services using Cloud-based IT resources. As long as malicious virtual machines could affect its host and other machines, the virtual machine placement is an important issues. In this paper, we outline an approach based on a combination of graph theory and agent based simulations to study epidemics in the cloud computing and control the transmission of infections. We present our results on the graph properties of virtual machines graph, as well as simulation-based dynamic analysis of infection spread in such networks. We identify a number of measures that are significant for understanding epidemics and for developing new strategies in virtual machines placement algorithm.