Research On Resource Management In Multimedia Cloud Computing

With the development of cloud technology,cloud computing is used as a distributed platform for multimedia services.However,multimedia service providers face many problems with resource management.How to optimize task scheduling and resource allocation under the strict QoS requirements is still a challenging task.Multimedia cloud computing mainly provides services in the form of virtual machines.Therefore,resource allocation problems include the placement of virtual machines.An efficient resource allocation mechanism is an unavoidable hot issue.In this paper,the task scheduling and resource allocation optimization,virtual machine placement,and resource monitoring strategy are researched as follows:(1)In order to allocate resources reasonably and minimize the task completion time,a new task management system is proposed.The task model and the task management based on queuing theory and the idea of combining ant colony algorithm in task scheduling strategy are elaborated in detail,and the function of dynamically expanding resources is added.In this paper,the improved ant colony algorithm is applied to the task scheduling system of dynamic resource allocation to further optimize the resource scheduling strategy.The simulation results show that the proposed scheme can effectively distribute tasks and schedule workload,shorten service response time and improve user satisfaction and resource utilization.(2)According to the diversity of multimedia resources,diversity of services and heterogeneity of server resources,a virtual machine placement algorithm is proposed.Firstly,according to the resource requirements of virtual machines and the supply of server resources,considering the QoS characteristics of multimedia services,a multi-objective programming model for resource allocation problem is established,and then the resource similarity and affinity are used to determine the resources between virtual machines and physical machines Relationship,and apply it to the virtual machine placement algorithm to find the optimal solution.Compared with the existing algorithms,experiments show that the proposed deployment scheme will not affect the quality of service,but also save costs.(3)Aiming at the problem of large amount of monitoring target in multimedia cloud platform and the amount of data spread affecting the network pressure,an adaptive acquisition interval and the intelligent switch between push and pull mode are proposed.First of all,the active monitoring and passive monitoring of the two monitoring modes are combined,and then control the switching of the monitoring mode according to the user's needs.At the same time,the collection interval and the number of transmission are controlled by the prediction algorithm.The experimental results show that the model reduces the update cost and meets the user's requirement of monitoring resource consistency.