Memory management method and related equipment

Abstract

The invention provides a memory management method and related equipment, which aims to actively perform memory defragmentation based on application scenarios and continuous memory demand prediction soas to meet the demands of different application scenarios for continuous memory, reduce the waiting time of memory allocation and improve the application running efficiency. The method comprises thefollowing steps: the terminal equipment acquires a switching probability of switching from a currently running first application scene to each second application scene in one or more second application scenes; determining a target continuous memory according to the continuous memory required by one or more second application scenes of which the switching probability meets a preset condition in theone or more second application scenes; and if the available continuous memory on the terminal equipment is smaller than the target continuous memory, before the terminal equipment is switched from the first application scene to any second application scene, the terminal equipment performs memory fragmentation, so that the available continuous memory on the terminal equipment is greater than the target continuous memory.

Description

technical field [0001] The present application relates to the field of computers, in particular to a memory management method and related equipment. Background technique [0002] The fragmentation of physical memory, that is, the discontinuity of memory pages, has always been one of the important problems faced by the operating system, and most of the memory used by general applications at runtime needs to be continuous memory. In order to solve the problem of physical memory fragmentation, the prior art usually uses a memory management algorithm, such as the defragmentation algorithm of Buddy system (partner system) in Linux, to organize the fragmentation (fragmentation) in the memory into continuous memory, so as to satisfy The memory requirements of the application. [0003] Existing memory management algorithms are mainly divided into two categories: synchronous memory defragmentation algorithms and asynchronous memory defragmentation algorithms. Among them, the synchr...

AI Technical Summary

Problems solved by technology

For example, the synchronous memory defragmentation algorithm triggers defragmentation when the system cannot allocate continuous memory for the current application. Therefore, it needs to wait for the system to release memory and sort out continuous memory before completing memory allocation. This will greatly increase the waiting time for memory allocation and affect the current application. application performance

The asynchronous memory defragmentation algorithm triggers memory defragmentation when the available continuous memory of the system is lower than the preset threshold, and stops when it reaches above the threshold. When the application requires a large amount of continuous memory, asynchronous defragmentation cannot meet the memory requirements of the application in time. , so as to enter synchronous memory defragmentation, which will also cause the problem of long memory allocation time and affect the operating efficiency of the application

Images