Task Scheduling Algorithm Based On Improved Particle Swarm Optimization Algorithm In Cloud Computing Environment

As the next generation application model of the Internet, cloud computing is also abusiness model. Facing with a large number of users, cloud computing need toprocess thousands of data and tasks in time, the level of cloud computing platformplays a decisive role in it. To a large extent, task scheduling algorithm affects theperformance of the cloud computing platform, so improving the task schedulingalgorithm becomes a research hotspot. The research of the task scheduling andresource allocation is few, and the existing task scheduling algorithms for cloudcomputing usually lay their attention on the pursuit of the shortest completion time,however, they are not well to take into account the cost of all the tasks for the pursuitof the shortest completion time. Because the cost of all the tasks is a important factorwhich can not be ignored. Different computing resources in cloud platform havedifferent costs, the use of excellent computing resources is higher, while the use ofcommon resources is lower. The cloud users usually make their choices based oncomprehensive consideration of the economic budget as well as the wait time. Tosolve the problem, A double fitness particle swarm optimization algorithm(DFPSO)based on the time and cost constraints is proposed in this paper, In the algorithm, Boththe complete time of all tasks and the cost of all tasks are used as schedulingobjectives. Through this algorithm, the proposed task scheduling not only shortenstotal task completion time and also costs less.It is an important issue for the cloud computing to how to reasonably allocate thecomputing resources and balance resources’ load. Existing task scheduling algorithmsfor cloud computing are not well to take into account the load balancing problem forthe pursuit of the shortest completion time. It may lead to the phenomenon ofunbalanced resources’ load. To solve this problem, a double-fitness particle swarmoptimization(DFPSO) based on resource pre-classification is proposed in this paper.In the new algorithm, the resources classifies through the information of measuring the ability of computing and communications, and then calculate the product of theexpected execution time of tasks in the resource and its corresponding resource class,which is regarded as another fitness function of task scheduling. The results generatedby this algorithm not only make the task completion time shorter, but also have ahigher utilization of system resources, which are taking into account the minimumexecution time and load balancing. The simulation shows that DFPSO is an efficienttask scheduling algorithm in the cloud computing by contrast with the conventionalparticle swarm optimization(PSO).