Resource Allocation Algorithm For Cloud Computing Based On QoS

As a new computing pattern, Cloud computing connects to lots of computers through Internet, which abstracts resources from hardware and software by use of open technology and standards, and provides to users appearing in the state of service. However, as a new thing, most parts of Cloud computing are still not solved well, such as resources scheduling. How to use limited resources and reasonable tactics to guarantee users’satisfactory and prove utilization is a difficult problem. This paper dose much researches about Cloud Computing and resources scheduling methods and summarizes these problems. Traditional scheduling methods usually judge performance by servers’utilization or responding time, but the ultimate aim of Cloud computing is satisfy user’QoS needs and whether the users can get the most benefits through the less cost. Obviously, traditional scheduling methods cannot do this. Under the circumstance of resources scheduling, because of commercial application, it should change from a system-centric to user-centric, as to improve customer satisfaction in the promise of ensuring system performance.As cloud-based services become more numerous and dynamic, resource provisioning becomes more and more challenging, we can’t measure performance of the algorithm from cost or time solely. To solve the above problem, a QoS constrained resource allocation algorithm is considered in the paper, in which we consider from user and resource provider. We must guarantee QoS of every task, at the same time, seek the promotion of utility. The Algorithm is different from resource scheduling algorithm of traditional cloud computing which focuses on system indicators, which applies to the characteristics of the commercialization of cloud computing. The main share of this article contains the following aspects:first, we consider the allocation of resources from consumers and resource providers, and then propose game utility function which contains some parameters, such as task completion time, task cost and task energy consumption. Second, in order to achieve the utility maximization of the global task, game-theory was introduced into the resource scheduling, which contains two steps:initial optimization and evolutionary optimization.After a reasonable analysis, this paper finishes a test by use of simulation software called CloudSim, which simulates the Resource allocation algorithm for cloud computing based on QoS, which contains cloudlets which limited by the maximum completion time、the largest spending and the maximum power consumption of resources. Finally, we increase task utility.In addition, the algorithm is suitable for multiple tasks competing for limited resources. The simulation results show that the resource allocation algorithm can reduce competition for resources among tasks, and improve the global utility value of all the tasks.