| Summary: | 碩士 === 國立東華大學 === 電機工程學系 === 99 === In the research of resource allocation, there are various significant issues, such as the maximum computing performance and the green computing which attract our attention recently. Therefore, how to accomplish tasks with the lowest cost has become an important issue when the resource on the earth is getting less and less. The goal of this research is to design an optimized resource allocation system composed of two sensing devices in cloud computing environment. Support vector regression (SVR) and genetic algorithm (GA) are applied in the proposed system. One of these two sensing device, called the application service monitor, is used to monitor the response time (RT) of each application service for obtaining collected data to predict the next RT. The other one, called the physical machine monitor, is used to monitor the utility rate and the remainder of resource in each physical machine. Furthermore, the prediction mechanism is performed precisely by using SVR, and the resource reallocation will be considered according to the operation of all virtual machine installed in physical machines.
A resource dispatch mechanism using genetic algorithms is proposed in this study to calculate the optimal reallocation of resources. The proposed scheme achieves an optimal configuration via reaching the agreement between the utilization of resources within physical machine monitored by physical machine monitor and Service Level Agreements (SLA) between virtual machines operator and cloud services provider. In addition, the mechanism can fully utilize hardware resources and maintain the overall performance in the cloud environment in the optimal situation.
|
|---|
