Research On Resource Scheduling Based On Particle Swarm Optimization Algorithm In Cloud Environment

As a popular technology in today’s IT sector,Cloud computing has been developed and researched by the world’s computer-related researchers and IT companies and has been widely used in various fields.Cloud computing can provide users with simple,convenient and inexpensive services through connecting the physical server using the network and composing a large-scale virtual resource pool by virtualization and some other technologies.As the popularity of the network,the users of cloud services are becoming more and more demanding on the quality of cloud services.Only low prices and good quality services are popular,however,Service providers not only provides high quality services but also wants to get more profits,so,a reasonable allocation of cloud services and scheduling appears to be more and more important.With the development of cloud computing,the use of cloud computing is not only the field of IT users,users are not limited to large enterprises,so it’s important for many non-IT companies and small businesses to choose the service.So,in order to solve the users’problem,the service agent can provide users with all the resources on the SaaS,PaaS,IaaS,after pulsing the service agent,the service agent can provide users with an intermediary service agent for the users to customize the most suitable cloud services for their business development.After that,the agent can not only improve the customer’s satisfaction but also reduce the cost of the users to use resources.At the same time,how to maximize the benefits of resource providers is also a focus of concern today.In order to meet the needs of cloud service users and satisfy the interests of resource providers,this paper proposes a resource scheduling algorithm based on particle swarm optimization under the constraint of Service Level Agreement.The experiment results show that the algorithm can improve profitability of the resource provider.However,the particle swarm algorithm is easy to fall into the local optimum.The resource scheduling scheme obtained by the Particle Swarm Optimization algorithm is also focalized,and the search space is small and the virtual machine is busy.In order to solve the problem that the particle swarm algorithm is easy to fall into the local optimum,the original particle swarm algorithm is transformed into a Multi-population Particle Swarm Optimization algorithm,which improves the search range of the optimal solution of particle swarm,so that the particle jumps out of local optimum.The results show that the scheduling algorithm can increase the profit of the resource providers,but also distribute the virtual machine evenly and improve the utilization rate of the resources.