Research On Task Scheduling And Virtual Machine Consolidation In Cloud Computing Environment

With the development of the Internet and mobile communication technologies,cloud computing provides users with the necessary resources and services through on-demand and pay-as-you-go services,becoming an essential service for the Internet.The uncontrollable randomness and massiveness of tasks submitted by users,the heterogeneity of resource nodes from data centers and the integration of task scheduling on hyperscale data center into virtual machine are creating unprecedented challenges for cloud service providers to overcome.Although the potential of cloud has been discussed for a number of years now,in both industrial and research communities,there has been little or no advancement in the field until recently.Cloud computing serves a large number of users.Different types of users have different requirements for services.Therefore,different types of users have different perceptions of service quality,and the scheduling policy without task type awareness can not make perfect adjustments to the task scheduling scheme according to different needs of users and current virtual machine resource status.Basically,cloud computing serves a large number of users.Since different types of users have different requirements for services,users have different perceptions of service quality.Therefore,scheduling policy without awareness of task type is incapable to make an idea adjustment to the task scheduling scheme according to different needs of users and current virtual machine resource status.In addition,virtual machine consolidation technology is considered to be an efficient load balancing and environmentally friendly energy conversation technique for cloud computing data center.However,the current approaches without considering the impact of virtual machine migration on future job scheduling and resource utilization efficiency bring great hidden danger.In this thesis,we survey existing algorithms related to task scheduling and virtual machine consolidation technology under computing cloud environment.Meanwhile,an optimized algorithm and data model are proposed.The main tasks as follows: Aiming at the problems existing in the above existing researches,after conducting in-depth research on task scheduling and virtual machine consolidation technology in cloud computing environment,the related algorithms and data models are proposed.Researched on the cloud computing architecture,resource scheduling methods and CloudSim simulation platform.A cloud computing task scheduling model corresponding to multi-objective QoS optimization,where constraint conditions and optimization goals aimed at requirement of three QoS indicators(load balancing,completion time and execution cost)are established,and a virtual machine integration model based on the awareness of resource utilization efficiency are proposed.Also,a few key concepts in later model such as the definition of user breach rate,power consumption status,load status and virtual machine deployment rationality by means of mathematical model are described.Since antlion optimization algorithm(ALO)is widely introduced to solve cloud computing scheduling problems,we optimize the adaptive steplength antlion optimization algorithm(ASALO)based on the task granularity characteristics of cloud computing.Moreover,the search step size is dynamically set according to the number of tasks.The simulation results show that the search speed and search accuracy are improved compared with PSO algorithm and ALO algorithm.We present a virtual machine consolidation algorithm associated with threshold value to take into account the efficiency of resource utilization.The simulation results show that the virtual machine consolidation algorithm triggered by the threshold reduces the service level agreement violation rate and energy consumption by reducing the optimal resource deviation coefficient of the system when considering the resource utilization efficiency.