Research And Implementation Of Job Scheduling Algorithm In Cloud Computing

Cloud Computing is becoming one of the hot target in commercial and scientific research institutions. Cloud computing is the development of grid computing and parallel computing and distributed computing. It is a new business computing model which provides on-demand packaged resources and services for client from internet and it adopts mature virtualization technology to package resources in data center. Job scheduling and resource allocation are two key technology in cloud computing. Cloud computing's commercial characteristic makes it focus on user's Quality of Service. And it's virtualization technology makes resource allocation and job scheduling significantly different from the previous parallel distributed computing.This paper has done a systematic and comprehensive research in cloud computing. It proposes an algorithm of job scheduling based on the Berger's Model which is a sociological theory for distribution justice. This algorithm adapts to the characteristics of cloud computing, which is commercialization and virtualization. It is different from traditional job scheduling algorithm which is focusing on the efficiency. It provides dual fairness constraints in cloud computing. First, It classifies the user's tasks by QoS and gives the description of task based on the Human Element Model. It creates general expectation according to the classification of tasks to constraint the resource selecting process. It also gives the method mapping the tasks and resources. Second, It defines the Justice Evaluation Function in cloud computing to justify the fairness of the resource allocation and do corresponding amending to the algorithm model. This function constraints the fairness of the cloud system. It is also provides a method based on the Human Element Model to process tasks with multiple expectations. Finally, this paper extends the CloudSim simulation tool to implement the job scheduling algorithm and recompiles it and implements the simulation with the recompiled CloudSim tool. It is proved to be valid by the simulation analysis.In addition, It has researched the load balance control in cloud computing basis on the Berger Model's balance theory. It provides a new direction for research of load balance control in cloud computing.