Enabling security-aware virtual machine placement in IaaS clouds

Abstract

Infrastructure as a Service (IaaS) facilitates the provisioning of virtual machines (VMs) in cloud computing platform for disjoint customers in a highly scalable, flexible, and cost-efficient fashion. However, introducing new VMs to a physical server where vulnerable VM already exists could lead to potential security risks to the new ones. Furthermore, even the physical server itself could be compromised by attackers through one of these vulnerable VMs. Therefore, VM placement could bring great impact over the security level of the whole cloud. In this paper, we first quantify the security risks of cloud environments based on virtual machine vulnerabilities and placement schemes. Based on our security evaluation, we present a novel VM placement algorithm that can minimize the cloud's overall security risks by considering the connections among VMs. According to the experimental results, our approach can greatly improve the survivability of most VMs and the entire cloud. The computing costs and deployment costs of our techniques are also practical.