Machine learning-based operating system scheduling method, device and device

Abstract

The invention discloses an operating system scheduling method and a device based on machine learning. The device comprises the following steps: classifying tasks into one or more task groups based onresource requirement characteristics by using an unsupervised clustering algorithm; the pre-established training model being used to adjust the resource allocation parameters of the task group, so that the hardware resource utilization efficiency and the degree to which the task meets the user expectations are optimized; the resource allocation parameter obtained by the training being used for scheduling and allocating resources for the process of the next cycle of the task group.

Description

Technical field [0001] The present invention relates to the field of computer technology, in particular to a method, device and equipment for operating system scheduling based on machine learning. Background technique [0002] The operating system manages the hardware resources of the computer. Modern operating systems allow multiple processes to share hardware resources in time sharing. The scheduling system is used to solve the problem of how to allocate hardware resources among multiple processes, that is, how to allocate the limited resources managed by the computer when and when How much to one process, when to allocate to another process, when to recover the resources allocated to one process. [0003] The scheduling system is the core module of the modern operating system. In the face of various complex application scenarios, how to allocate the limited system resources as reasonably as possible to the processes that need to be scheduled is the goal of the scheduling system....

AI Technical Summary

Problems solved by technology

However, in actual operation, due to the complexity of the system, the priority setting or task category setting is often improper, resulting in low efficiency of system resource allocation, and finally causing tasks to be scheduled by the operating system inconsistent with user needs. In severe cases, some tasks may Occupying the CPU for a long time, other tasks cannot be executed in time

Even if the user sets more appropriate parameters for specific tasks, due to too many tasks, the number of parameter combinations is huge, and it is impossible to try them one by one. The parameter combinations set by users for all tasks are often not relatively reasonable, so that it is impossible Get the most out of your computer

[0007] There is another problem in the scheduling method of the related technology, that is, for some special scenarios, the pre-designed scheduling strategy and scheduling parameters often cannot play a good role

[0008] In summary, on the one hand, the scheduling method of the related technology requires the user to set the scheduling policy and scheduling parameters for each specific task, and in the case of a large number of tasks in the system, the user either cannot set a reasonable scheduling method for one or some specific tasks. Scheduling parameters (such as priority or task category) lead to low system resource allocation efficiency, or the user sets reasonable parameters for specific tasks but the combination of these parameters cannot give full play to the performance of the computer. On the other hand, the scheduling of related technologies The method is not flexible enough, and the preset scheduling strategy and scheduling parameters cannot play a role in some special scenarios

For the above technical problems, no effective solution has been proposed yet

Images