A parallel program task processing method and device thereof

Abstract

The invention discloses a parallel program task processing method and a device thereof, comprising assigning an unprocessed task to each processing core according to a first preset rule from the current unprocessed task; when the processing core processes the currently assigned task , according to the data inheritance between tasks, select a task related to the task currently processed by the processing core as a follow-up task; when there is no follow-up task, select a task from the current unprocessed task according to the second preset rule Assign an unprocessed task to the processing core; when the processing core completes the processing of the currently assigned task, store the obtained task data in the cache corresponding to the processing core, and process the subsequent task as the currently assigned task of the processing core; until all tasks are processed. Several related tasks in the present invention use the same cache, so the processing efficiency is high.

Description

technical field [0001] The invention relates to the technical field of process processing, in particular to a parallel program task processing method and a device thereof. Background technique [0002] In the current parallel program task processing, the commonly used method is the orderly and irregular parallel program task processing method. The parallel task processing process under this rule has three characteristics. The first is that the tasks must follow the overall or partial order (basically time sequence); the second is that the task may have data dependencies that are not known in advance; the third is that the task may dynamically create subtasks and plan to run them in the future. [0003] In this way, since multiple processing cores process in parallel, and the tasks are assigned to each processing core in sequence according to the time sequence, it is easy for several related tasks to be processed by different processing cores, which leads to subsequent tasks ...

AI Technical Summary

Problems solved by technology

[0003] In this way, since multiple processing cores process in parallel, and the tasks are assigned to each processing core in sequence according to the time sequence, it is easy for several related tasks to be processed by different processing cores, which leads to subsequent tasks to be processed It is necessary to read the required data from the cache corresponding to the processing core of the previous task before performing the task, which is inefficient (for example, task 2 is related to task 7, and the processing core corresponding to task 2 needs to be processed first when task 7 is processed. Cache out the data of download task 2)

Images