Research On System Management And Optimization Of Mobile Cloud Computing

With the rapid development of mobile Internet and cloud computing,mobile cloud computing has become the latest promising research direction.The rapid improve in performance and processing capabilities of intelligent mobile devices,as well as the application of distributed processing technologies such as Hadoop and Map Reduce on mobile devices,have led mobile devices gradually starting to act as mobile cloud server nodes and participate in high-performance computing to handle large amounts of data.However,mobile devices themselves have mobility,mobile network coverage also exists gaps,and different devices show different stability and processing capacity when dealing with tasks.Besides,different tasks have different needs in the various resources,those have brought challenges to the unified management and distribution of mobile device resources.Thus,the resource management and task allocation of mobile cloud computing are studied in depth in this paper.The contents and innovations of this paper are as follows:(1)Select a proxy server-based mobile cloud computing system architecture,using proxy to connect mobile devices and clouds.And design the functional modules including monitor,resource manager,task manager and scheduler in the proxy for reliable resource management and efficient task allocation.(2)A mobile resource clustering method based on improved K-means algorithm PKMC_RM is proposed.Considering the mobility,reliability and resource constraints,the mobile device resources are clustered according to their stability and mobility.Each mobile device cluster performs tasks with different resource requirements according to its processing capability.The detailed design and implementation of PKMC_RM are given,including the selection and standardization of input parameters,the calculation of distance,the selection of initial cluster center,the search of newcluster center and the improvement of the rule of continuous search.The experiment proves that PKMC_RM can achieve better clustering effect to meet the reliability requirements of resource clustering management.(3)A multi-objective optimization task allocation method based on ant colony algorithm MOAC_TD is proposed,and the optimal allocation of tasks is completed in two stages.The design and implementation of MOAC_TD are given,including the tasks clustering process,selecting load balancing and minimum power consumption as two optimization goals,the design of fitness function,the improvement of pheromone update rules and the design of state transition probability,and using Pareto domination relationship to find the best solution in each iteration.The experiment shows that MOAC_TD can maintain the balance of system load and have better energy saving effect on the basis of satisfying the tasks request to resources.