Task scheduling method and system

Abstract

The invention provides a task scheduling method and system. The method includes: determining the candidate task set of each processor in a system according to task I / O resource description vectors to allow the candidate task set of each processor to include tasks which are appointed to the processor, reaches an I / O busy stage and has I / O resource demands. The method further includes: for the processors whose I / O demands are larger or smaller than the I / O bandwidth thresholds of the processors, respectively hanging up or restoring the tasks in the candidate task sets of the processors according to the sensitivity of the task I / O resource description vectors. The method is applicable to various applications sensitive to I / O competition, and task executing efficiency and system resource utilization rate are increased.

Description

technical field [0001] The present invention relates to the field of computer parallel technology, and more specifically, to a task scheduling method and system. Background technique [0002] In recent years, the on-chip multi-core architecture has made great progress. At the same time, based on the powerful computing capabilities provided by the on-chip multi-core architecture, cloud computing has also begun to become the first choice for many manufacturers' IT infrastructure. In order to provide this powerful computing capability, cloud computing providers organize together a huge number of on-chip multi-core processors to form a cloud computing data center. In this case, the efficiency of the data center and the utilization of hardware resources become a very critical issue. [0003] Research data shows that I / O resource conflict is one of the important factors affecting the efficiency and resource utilization of data centers. With the expansion of scale, the data cente...

AI Technical Summary

Problems solved by technology

However, this method of real-time local I / O throttling and global coordination has the same problem as the previous method, that is, there is a lack of regulation for jobs with the same priority. When there is I / O between jobs with the same priority When resources compete, it will cause performance loss and reduce resource utilization

[0006] In summary, the existing task scheduling methods have certain limitations, and these methods only focus on I / O flow-intensive applications, ignoring other types of applications that are sensitive to I / O competition (such as graphics and image processing, backup, etc.) and other input-dependent applications)

Images