Dynamic partitioning method, system, equipment and medium

Abstract

The embodiment of the invention provides a dynamic partitioning method, system and device and a medium. The method comprises the following steps: acquiring a first partition level of a first partition; wherein the system processing resource has a hierarchical structure, the hierarchical structure comprises layers corresponding to different partition levels, and the first partition is located in a first layer in the hierarchical structure; determining working parameters of a plurality of control groups contained in the first partition; determining a second partition level according to the working parameters of the plurality of control groups; and when the second partition level is not matched with the partition level, adjusting the plurality of control groups into the partitions contained in the layer corresponding to the second partition level. When dynamic partitioning needs to be carried out, adjustment can be carried out among layers corresponding to different partition levels of the same partition branch according to the processor pressure of the partition level corresponding to the first partition. And when the cross-affinity domain adjustment demand is carried out, the control group is taken as a division unit to carry out overall migration between different partitions of the layer where the same partition level is located, so that the performance of the overall processor can be effectively improved.

Description

technical field [0001] The present application relates to the field of computers, in particular to a dynamic partition method, system, equipment and media. Background technique [0002] With the development of embedded operating system technology, the demand for the processing capability of the operating system is also gradually increasing. For example, multi-processors (CPU) will be used to achieve a corresponding improvement in processing performance. [0003] In order to improve the processor utilization rate of the whole machine, a control group (control group, Cgroup) is configured to share the processor, and the deployment density is increased to fill the processor fragments. However, after the deployment density of this shared processor reaches a certain level, processor starvation occurs, such as cache and TLB competition leading to additional processor overhead. In addition, partitioning the processors of the whole machine according to NUMA nodes, and performing dy...

AI Technical Summary

Problems solved by technology

However, after the deployment density of this shared processor reaches a certain level, processor starvation occurs, such as cache and TLB competition leading to additional processor overhead.

In addition, partitioning the processors of the whole machine according to NUMA nodes, and performing dynamic partitioning work based on task migration on a single node, due to the high affinity, the efficiency is very low when tasks need to be migrated between partitions, which will cause processor reduced performance

Images