Hybrid Chemical Reactive Optimization And Tabu Search For Cloud Workflow Scheduling

Abstract:Cloud computing is a new method that provide server through the internet with the development of network. Cloud computing is the combination with these old techniques, include distribute computing, parallel computing, utility computing, network storage, virtualization and load balancing. As the development of Cloud computing, more and more researchers come to research the workflow scheduling for cloud.Oriented-QoS Workflow Scheduling needs to be improved in the management of workflow for cloud. There need some methods or some schedule policies to meet users’quality require, such as minimize the total cost of execution or minimize total execution time and balance the load on resources used while meeting the dead-line constraints of the application. Combined with the research status, in this paper, we try to use a new method to solve the problem of minimize total cost.To address this problem, we study the existing intelligent algorithm. Chemical Reaction Optimization is a new meta-heuristic method that is put forward in recent years. This algorithm is proved its’performance in QAP, Workflow Scheduling in Grid computing. We developed a new hybrid algorithm which combined Tabu search and Chemical Reaction Optimization. This new method gets the global exploring ability from CRO and local exploring ability from Tabu search, through the test functions that we proved its’exploring ability of the solution space. We design and implementation the molecular structure and the neighbor function of TS.In this thesis, we compared TSCRO, PSO and BRS, and the results show that it can ensure the resource not overloading and make the total cost minimize, this result makes TSCRO using in other problems in cloud be easier.