Garbage collection space management method, device and computer-readable storage medium

Abstract

The application discloses a garbage collection space management method, device and computer-readable storage medium. Wherein, the method includes pre-constructing multiple block recovery selection modules, and each block recovery selection module is divided into multiple inefficiency intervals with different inefficiency ranges. When it is detected that there is a target data block full of data, the target data block is migrated to the corresponding inefficiency interval in the target block recovery selection module according to the inefficiency of the target data block, and every first preset time based on the current inefficiency will be The target data block is adjusted to the corresponding inefficiency interval; when there is a block recycling selection module with the highest inefficiency interval that meets the garbage collection conditions, the data block is selected from each block recycling selection module according to the order of the data block usage time from old to new to be garbage Recycling, in the process of garbage collecting data blocks, it is realized that data blocks with more invalid data can be selected to reduce the amount of migrated data, and at the same time, invalid migration can be reduced to reduce system overhead.

Description

technical field [0001] The present application relates to the technical field of all-flash storage arrays, in particular to a garbage collection space management method, device, and computer-readable storage medium. Background technique [0002] In an all-flash storage array, in order to reduce the write amplification of the SSD (Solid-state disk) in the array and prolong the service life of the SSD, the storage array software usually handles the write business of the host in a log-structured redirected write mode , so as to convert the random small IOs generated by the host write business into continuous large IOs and write them into the SSD. In this way, after the host overwrites, garbage data will be generated, and garbage collection is required to reclaim the storage pool space containing invalid data in time. The space area occupied by garbage data cannot be reused to receive host data until it is reclaimed by garbage collection. [0003] However, the granularity of s...

AI Technical Summary

Problems solved by technology

Therefore, it is best to select a block containing a large amount of invalid data for recycling to reduce the amount of data to be migrated. However, because the storage pool has a large capacity and there are many blocks, it is not easy to find a block containing a large amount of invalid data.

In addition, if the data to be migrated is the data just written by the host, these data may soon become garbage data, thus resulting in invalid relocation, doing useless work and wasting storage array resources

Images