A Mobile Computing Model With Real - Time Performance In Mobile Cloud Environment

Mobile cloud computing is a combination of mobile communication and cloud computing. With the development of mobile cloud computing, an increasing number of mobile applications appear on the Internet,people’s demands for computing experience get higher.How to meet the demands on the condition that mobile devices are still resource-limited and mobile networks are still low-bandwidth and unreliable is a problem to be solved.Realtime performance is a high-demanding stuffPeople want to get their computed results within the stipulated time.Mobile cloud computing usually studies on the computation migration and computation offloading,we put forward a solution with realtime performace about computation migration in the mobile environment based on the existing researches.The specific work includes:task partitioning,computation migration model,multitasking and simulation experiments.The paper’s primary work as follows:Firstly,we make a comparison between the existing methods about the task partitioning and then assess the advantages and disadvantages,we give time property to the compute nodes and present RT-Partition algorithm with real time characteristics. The optimal solution is obtained by comparing the total time consumption of different migration orders.In the next place, this paper analyzes and compares the existing mobile computing task migration models and frameworks and assesses its advantages and disadvantages,then we give our migration model with realtime performance on the base of our task partition algorithm,finally we evaluate our migration model using defined evaluate model.Moreover,we study the scheduling theories in open real-time environment and solve multitasking in computation migration based on the improved real-time server DyCBS.In the end,we built our simulation platform MobilecloudSim and the migration model was proved effective by multi-group experiments.