Doctor of Philosophy (Ph.D.)
As cloud computing has become a trend in the computing world, understanding its security concerns becomes essential for improving service quality and expanding business scale. This dissertation studies the security issues in a public cloud from three aspects. First, we investigate a new threat called power attack in the cloud. Second, we perform a systematical measurement on the public cloud to understand how cloud vendors react to existing security threats. Finally, we propose a novel technique to perform data reduction on audit data to improve system capacity, and hence helping to enhance security in cloud. In the power attack, we exploit various attack vectors in platform as a service (PaaS), infrastructure as a service (IaaS), and software as a service (SaaS) cloud environments. To demonstrate the feasibility of launching a power attack, we conduct series of testbed based experiments and data-center-level simulations. Moreover, we give a detailed analysis on how different power management methods could affect a power attack and how to mitigate such an attack. Our experimental results and analysis show that power attacks will pose a serious threat to modern data centers and should be taken into account while deploying new high-density servers and power management techniques. In the measurement study, we mainly investigate how cloud vendors have reacted to the co-residence threat inside the cloud, in terms of Virtual Machine (VM) placement, network management, and Virtual Private Cloud (VPC). Specifically, through intensive measurement probing, we first profile the dynamic environment of cloud instances inside the cloud. Then using real experiments, we quantify the impacts of VM placement and network management upon co-residence, respectively. Moreover, we explore VPC, which is a defensive service of Amazon EC2 for security enhancement, from the routing perspective. Advanced Persistent Threat (APT) is a serious cyber-threat, cloud vendors are seeking solutions to ``connect the suspicious dots'' across multiple activities. This requires ubiquitous system auditing for long period of time, which in turn causes overwhelmingly large amount of system audit logs. We propose a new approach that exploits the dependency among system events to reduce the number of log entries while still supporting high quality forensics analysis. In particular, we first propose an aggregation algorithm that preserves the event dependency in data reduction to ensure high quality of forensic analysis. Then we propose an aggressive reduction algorithm and exploit domain knowledge for further data reduction. We conduct a comprehensive evaluation on real world auditing systems using more than one-month log traces to validate the efficacy of our approach.
© The Author
Xu, Zhang, "Understanding Security Threats In Cloud" (2016). Dissertations, Theses, and Masters Projects. Paper 1477067957.