Secure virtual machine placement in infrastructure cloud services

• Main Authors: NATU, Varun, TA, Nguyen Binh Duong
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2017
• Subjects: Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/4833
• Institution: Singapore Management University

With increasing evidence on the breadth and potency of side-channel attacks based on resource sharing, virtual machine (VM) placement in infrastructure cloud services, such as Amazon EC2, has become an important security problem. In this paper, we present a new approach to this problem by letting users specify their own trust profiles, and then incorporating such information into VM placement decisions. The objective is to ensure safe VM placement while aiming to maximize physical resource utilization in data centers. We propose and implement four different solutions to the problem, namely Graph Coloring, Bin Packing, Linear Programming and Genetic Algorithm. The proposed algorithms have been thoroughly evaluated on various real world datasets and different trust profiles. We found that, by including security requirements within VM placement algorithms, the loss of efficiency was as little as 2-3% in certain trust profiles and datasets. In addition, simple greedy approaches were found to be more scalable; and performed much better in terms of efficiency loss than more complicated optimization methods.