Research On Resource Scheduling And Energy Consumption Management In Cloud Data Center

Because of the cloud data center includes a variety of physical devices,network equipment and large amount of resource management.In many case of a static resource management will appear some phenomenon that a few physical device under the heavy load and most physical device under the light load or idle.This phenomenon will lead to low utilization rate and cause unnecessary overhead.Therefore,achieving the goal of high utilization rate of cloud data center resources and meeting the service requirements of users are of great significance to solve the currently existing challenges.This paper analyzes the characteristics of cloud computing resource scheduling,proposes a multi-factor-based host overload detection strategy and an improved particle swarm optimization algorithm for the dynamic migration of virtual machines.The main work is as follows:(1)This paper analyzes the research situation and importance of cloud computing in government,enterprises and universities at home and abroad,and introduces the theoretical research status of the virtual machine scheduling problems of cloud computing data centers and some problems encountered in their development.(2)The system introduces the basic concepts of cloud computing technology and related technical characteristics,and in-depth study and research on cloud computing virtual machine consolidation technology.(3)In-depth study of the cloud computing simulation platform CloudSim I used,including its system architecture and various strategies used inside,let me understand the operation of the cloud computing data center at the code level.(4)A dynamic threshold host overload detection strategy based on the combination of CPU,memory,and bandwidth resources is designed for the cloud data center physical machine.The virtual machine maximum utilization selection method is designed to reduce the number of unnecessary virtual machine migrations;the low utilization threshold host underload detection method enables service level agreements to meet the needs of different users.The simulation platform was used to verify the effectiveness of our proposed strategies.A good solution to the process of virtual machine dynamic consolidation: a,when to choose the virtual machine migration;b,choose which virtual machine to migrate,these two main issues.(5)The particle swarm algorithm update formula is improved,and the fitness value calculation function is designed to suit the practical problem of virtual machine scheduling.With the appropriate coding,an optimization algorithm and a mixture of Gaussian and Cauchy mutations are added to the local search,so that the particle swarm algorithm can be applied to the allocation of virtual machines.The improved particle swarm algorithm is devoted to the common concern of resource utilization and service level agreements,so that it can solve the optimal solution under various resource constraints in a short period of time,making the virtual machine better deployed and greatly reducing data center energy consumption costs.Finally,it solves the problem of virtual machine migration to which host in the virtual machine dynamic integration problem.