A garbage collection method of a storage system and a related device

Abstract

The invention discloses a garbage collection method of a storage system, comprising the following steps of: obtaining IO coverage states corresponding to IO characteristics and IO characteristics, carrying out machine learning and training treatment according to the IO characteristics and IO coverage states, and obtaining a coverage probability prediction model; according to the coverage probability prediction model, respectively carrying out prediction processing on each large block space to obtain a plurality of coverage probabilities; marking a large block space whose coverage probability is smaller than a preset coverage probability as a large block space to be recovered; carryingout garbage collection treatment on all large chunks of space to be collected. Through the prediction modelof machine learning training to determine whether the effective data in the large space will soon become garbage data, so as to avoid the garbage collection treatment of this large space, improve theIO performance of the IO storage system, and avoid the waste of IO performance. The present application also discloses a garbage collection system, a server and a computer-readable storage medium, which have the above beneficial effects.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a garbage collection method for a storage system, a garbage collection system, a server, and a computer-readable storage medium. Background technique [0002] With the continuous development of information technology, more and more data are stored in the Internet, and in order to improve the efficiency of data storage, an AFA (all-flash storage) array appears. AFA arrays use SSD (solid-state drive) hard disks for storage. Due to the write characteristics of SSD itself and the limitation of the number of erases and writes, discrete data is usually aggregated and rewritten to realize garbage collection of large spaces. In order to efficiently use the SSD hard disk. [0003] Generally, the garbage collection method provided by the prior art is to count the total amount of garbage data in each large space, select the large space with the most garbage data as the larg...

AI Technical Summary

Problems solved by technology

[0004] However, in the existing technology, when the valid data in a large space is migrated to a new space, it becomes garbage data, which not only fails to achieve the effect of the garbage collection method, but also wastes the IO of the storage system for migrating valid data. (read and write operations), which affects the performance of the host and also affects the service life of the SSD hard disk

Images