2018/03/14 by Ahmed H. Anwar, Anwar, Ahmed H., George Atia +3 · 1 citation
Computer Science · #Blockchain Technology Applications and Security #Computer Science and Game Theory (cs.GT) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Network Security and Intrusion Detection #Security and Verification in Computing
paper · pdf · doi:10.48550/arxiv.1803.05542
openalex publication_date 2018/03/14 · openalex created_date 2022/10/06 · openalex updated_date 2026/07/28
In a multi-tenant cloud, a number of Virtual Machines (VMs) are collocated on\nthe same physical machine to optimize performance, power consumption and\nmaximize profit. This, however, increases the risk of a malicious VM performing\nside-channel attacks and leaking sensitive information from neighboring VMs. To\nthis end, this paper develops and analyzes a game-theoretic framework for the\nVM migration timing problem in which the cloud provider decides \when to\nmigrate a VM to a different physical machine to reduce the risk of being\ncompromised by a collocated malicious VM. The adversary decides the rate at\nwhich she launches new VMs to collocate with the victim VMs. Our formulation\ncaptures a data leakage model in which the cost incurred by the cloud provider\ndepends on the duration of collocation with malicious VMs. It also captures\ncosts incurred by the adversary in launching new VMs and by the defender in\nmigrating VMs. We establish sufficient conditions for the existence of Nash\nequilibria for general cost functions, as well as for specific instantiations,\nand characterize the best response for both players. Furthermore, we extend our\nmodel to characterize its impact on the attacker's payoff when the cloud\nutilizes intrusion detection systems that detect side-channel attacks. Our\ntheoretical findings are corroborated with extensive numerical results in\nvarious settings.\n