| dc.description.abstract | Since modern data centers have been significantly scaling up in capacity in past decades, it is demanding to curtail energy consumption of virtual-machine-powered data centers. Cloud computing has radically changed the landscape of computing, storage, and communication infrastructures and services. Cloud computing's benefits encompass on-demand capacity, low cost of ownership, and flexible pricing. In the first part of this dissertation I propose a frequency-aware management strategy, which controls dynamic power and static power of processors running virtual machines in data centers. Unlike existing dynamic voltage and frequency scaling schemes, my strategy simply incorporates frequency requirements rather than task execution times. This salient feature is practical because task execution times in a raft of real-world applications are unknown in a priori. I build a frequency-aware model to derive an optimal frequency ratio that minimizes processors’ energy consumption. With my model in place, the energy efficiency of a datacenter can be maximized by adjusting the processor’s frequency to meet the optimal frequency ratio. I design a management approach to judiciously adjust frequency ratio to conserve energy without violating the frequency requirements imposed by virtual machines. After analyzing the correlations between frequency ratio and energy consumption, I show that a small static power proportion gives rise to high energy-saving performance. The results demonstrate that my model lays out a solid theoretical foundation catering to the development of power management software in DVFS-enabled clouds.
Besides the energy consumption, security issues coupled with resource allocations in cloud computing remain a challenging problem to be tackled by the industry and academia. While moving towards the concept of on-demand services and resource pooling in a distributed computing environment, security is a major obstacle for this new dreamed vision of computing capability. In the second part of the dissertation study, I articulate novel energy-aware scheduling policies customized for virtual machines running on clouds, in which service-level agreements (SLAs) are fulfilled. After addressing security concerns in cloud computing, I advocate for a research roadmap towards future security-aware energy management in clouds. I propose a high-level design for a security- and frequency-ware DVFS model or SF-DVFS, which orchestrates security services, security overhead analysis, and DVFS control green cloud computing systems. I delve into the main technical challenges associated with the proposed SF-DVFS model. To solve this multi-objective problem, I design a Secure and Economical DVFS-enabled Scheduling Policy with NSGA-II-SER (Non-dominated Sorting Genetic Algorithm II with Security and Energy Requirement) for Clouds.
Blockchain is an ideal privacy protection technology characterized by decentralization, transparency, data security, and system autonomy. As the last project in this dissertation research, I navigate leading-edge energy saving and privacy protection techniques for clouds. Next, I investigate privacy controls in blockchain systems. Inspired by modern blockchain and cloud computing techniques, I elaborate on a research roadmap towards future energy-aware privacy protection mechanisms in clouds. In a case study, I design a blockchain-based VM consolidation framework accompanied by the DVFS (Dynamic Voltage and Frequency Scaling) technique to offer energy savings and privacy controls in clouds. I expect that the roadmap will open up potentials to develop energy-efficient blockchain-based cloud computing platforms. | en_US |
