Research On The Task Scheduling Algorithm In Cloud Computing Environment

Cloud computing is a new calculation model with obvious Internet features.It wills traditional computing methods that only aim at specific areas or specific users to commercialize and popularize.At present,well-known universities and scientific research organizations around the world have invested in the study of cloud computing.In the complex cloud environment,the rationality of cloud task scheduling is directly related to the operating performance of the cloud computing system.Therefore,the study of cloud task scheduling algorithm plays a significant role in the development of cloud computing.In order that cloud computing systems have better comprehensive performance and ensures customer’s quality of service.The cloud task scheduling algorithm must be optimized.How can reasonably allocate the "cloud" end task in the cloud computing system? In order to solve this problem,the following research works of task scheduling algorithm in the cloud computing environment have been done in the paper.(1)An improved genetic algorithm for cloud task scheduling is proposed.The algorithm is based on traditional genetic algorithm to join the dual elite reservation strategy,and according to performance of virtual machine resources under the cloud environment to introduce the concept of virtual machine relative fitness.Random mutation operation of traditional genetic algorithm is improved to the target mutation operation,which makes the algorithm quickly converge to the optimal solution.Simulation experiments show that the algorithm can reduce the task completion time and balance the virtual machine load,is a feasible cloud task scheduling algorithm.(2)Another improved simulated annealing cloud task scheduling algorithm with multi-dimensional QoS constraint attributes is proposed by summarizing shortcomings of the cloud task scheduling improved genetic algorithm on the customer service quality.Aiming at the established mathematical model of cloud task scheduling with Qo S constraint attributes,the algorithm introduces the multi-dimensional QoS constraint time greedy strategy to generate the initial solution.It implements the improved simulated annealing process,which is always under the user’s QoS constraint to search the best allocation scheme of cloud task scheduling.Simulation experiments indicate that the proposed algorithm can meet the users’ QoS needs,decrease the task completion time and speed up its execution.It is an efficientalgorithm which can take account of users and cloud service providers at the same time.