dc.description.abstract | Cloud computing is not a brand new technology but rather a kind of conceptual innovation and change. Cloud computing integrates conventional computing resources, network resources, storage resources and related software systems into internet services, typically in a “pay-as-you-go” manner. However, it has become a common concern for the cloud service providers to balance resource usage and system performance. That is, consolidating computing tasks to a smaller set of working machines usually implies less operational expense (such as power consumption) for the cloud service providers. Meanwhile, it also may increase the workload of the set of the working machines and consequently decrease the performance of the whole system. For a cloud platform of limited resources that faces a peak workload, it is critical to use a good resource allocation strategy to achieve better load balance that yields better system performance. This thesis presents a resource allocation strategy, which is implemented on a new cloud platform, namely “SAMEVEDStack.” The concept of SAMEVEDStack is derived from a platform, “SAMEVED” (System Architecture for Managing and Establishing Virtual Elastic Datacenters), which is developed by the Department of Computer Science and Information Engineering, National Central University, Taiwan. The SAMEVEDStack is a non-commercial experimental cloud platform. It is implemented on the architecture of OpenStack, and designed for conducting on-line network and computer security experiments. The experimental results show that, the proposed resource allocation strategy outperforms three prior popular resource allocation strategies on our SAMEVEDStack cloud platform. | en_US |