Optimization of Task Scheduling and Resource Allocation on Cloud Environment
| Autor: | Jia-cen Fang, 方嘉岑 |
|---|---|
| Rok vydání: | 2012 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 100 In recent years, more and more demand need for cloud computing, the main problem of cloud computing is how to shorten the finish time and reduce the cost of the job. Allocation of task and resource is important, we need to take many factor into consideration, such as execution speed of resource, data size, rent cost of resource, execution time and waiting time, etc. We simulated a model for our research, adopted the execution cost, communication cost and waiting cost, combined differential evolutionary algorithm and Taguchi method to find the optimal solution. We also use multi-objective concept to find non-dominated solution.Parameter can be adjusted dynamically, for example, solution with shorter time but higher price, solution with longer time but cheaper price, or balance. Our purpose is to find the best non-dominated solution of the job scheduling. In our research, we compared differential evolutionary algorithm combind with taguchi method, differential evolutionary algorithm and genetic algorithm .We found that differential evolutionary algorithm combind with taguchi method, we can get better convergence and search result. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|