| The scheduling algorithm in cloud computing has always been the research hotspot.Until now,there are many studies on single DAG and the scheduling algorithm about single DAG is also very mature.Because of the arrival time of the task is random,we can't use the single DAG scheduling algorithm to schedule multi-DAGs.In addition,most of the DAG scheduling algorithms only take the performance of the processor,bandwidth,the cost of the task and other factors into consideration,but regardless of the effect of reliability on tasks' execution time.However,due to the large scale of the cloud computing system and the complexity of the software,the reliability of the task also become an important factor to makespan of the task.In order to improve the reliability of tasks and to schedule multiple-DAGs which arrival randomly,this thesis proposes two algorithms,the High Reliability Scheduling Algorithm Based on Task Duplication(HRSA)and the High Reliability Scheduling Algorithm for Multiple DAGs(HRSAMD).The HRSA algorithm can replicate the tasks and fairly scheduling them on a set of processors,by this way,the reliability of tasks is improved.Compared with the existing lower limit of task duplication strategy algorithm,HRSA algorithm is more equitable in the allocation of processors.This algorithm can not only meet the user's requirements of reliability,but also schedule the DAGs more effectively.The HRSAMD algorithm is a dynamic scheduling algorithm.This algorithm can dynamically deal with multiple DAGs,which can arrive at any time.The HRSAMD algorithm can split the DAGs into individual jobs,and then the HRSA algorithm is used to schedule the multi-copy of jobs,so as to improve the reliability of the DAGs and meet the user's reliability requirements.The simulation results show that the HRSA algorithm can complete same tasks set in less time than the existing algorithms,and the HRSAMD algorithm can effectively reduce makespan of DAGs while satisfying the user's reliability requirements. |
PDF Full Text Request