METHOD FOR SCHEDULING APPLICATION TASK BASED ON BIG DATA BEHAVIOR, SERVER AND STORAGE MEDIUM

摘要

Disclosed are a method for scheduling an application task based on a big data behavior, a server and a storage medium. The method comprises: collecting current resource state information of a server; if it is determined according to the resource state information that the current resource of the server is sufficient, starting a plurality of threads, and calling the plurality of threads to execute a plurality of tasks concurrently, no dependency relationship existing among the plurality of tasks, and the dependency relationship comprising at least one of the following items: a sequential order in the execution sequence of the tasks, a dependency relationship of simultaneously triggering in parallel, or the relationship that the execution of one task depends on the completed execution of the previous task; and scheduling the execution sequence among the tasks without the dependency relationship in real time according to the historical records of the execution durations of the tasks and the resource state information, and traversing all the tasks in the plurality of tasks until all the tasks are executed. The task execution efficiency can be improved, and the service influence caused by the fact that the tasks are unfinished is reduced.