Due to the large and frequent static data interaction between the Electric Information Acquisition System and the external business systems,researching on using limited server sources to do an efficient task schedulin...Due to the large and frequent static data interaction between the Electric Information Acquisition System and the external business systems,researching on using limited server sources to do an efficient task scheduling is becoming one of the key technologies of the unified interface platform.The information interaction structure of the unified interface platform is introduced.Task scheduling has been decomposed into two stages,task decomposition and task combination,based on the features(various types and dispersed)of large static data.The principle of the minimum variance of the subtasks data quantity is used to do the target task resolving in the decomposition stage.The thought of the Greedy Algorithm is used in the task combination.Breaking the target task with large static data into serval composed tasks with roughly same data quantity is effectively realized.Meanwhile,to avoid the situation of the GA falling into the local optimal solution,an improved combination method has been put forward.Moreover,the new method creates more average composed tasks and making the task scheduling more effective.Ultimately,the effectiveness of the proposed method is verified by the experimental data.展开更多
文摘Due to the large and frequent static data interaction between the Electric Information Acquisition System and the external business systems,researching on using limited server sources to do an efficient task scheduling is becoming one of the key technologies of the unified interface platform.The information interaction structure of the unified interface platform is introduced.Task scheduling has been decomposed into two stages,task decomposition and task combination,based on the features(various types and dispersed)of large static data.The principle of the minimum variance of the subtasks data quantity is used to do the target task resolving in the decomposition stage.The thought of the Greedy Algorithm is used in the task combination.Breaking the target task with large static data into serval composed tasks with roughly same data quantity is effectively realized.Meanwhile,to avoid the situation of the GA falling into the local optimal solution,an improved combination method has been put forward.Moreover,the new method creates more average composed tasks and making the task scheduling more effective.Ultimately,the effectiveness of the proposed method is verified by the experimental data.
