Data migration method and device

Abstract

The invention discloses a data migration method and device. The method comprises the following steps: in response to a received data writing request, writing a file requested to be written by the data writing request into a high-level storage pool; in response to the fact that the file requested to be written has been written into the advanced storage pool, determining whether the file requested to be written by the data writing request is a large file or a small file at the metadata server; writing metadata of a large file into a large file table in response to the fact that the file requested to be written is the large file, and writing metadata of a small file into an index table in response to the fact that the file requested to be written is the small file; and in response to the judgment that the condition for executing data migration is met, migrating the large files in the large file table from the high-level storage pool to the common storage pool according to the metadata sequence of the large files by the metadata server. The data file migration efficiency can be improved, a high-level storage medium is prevented from being worn by writing, the user experience and the user satisfaction degree are improved, and the product competitiveness is improved.

Description

technical field [0001] The present invention relates to the field of metadata loading, and more specifically, to a data migration method and device. Background technique [0002] In the existing technology, files are preferably stored in advanced pools to improve read and write performance. Since the advanced pool is composed of high-cost storage media such as SSD (solid state drive), and a file is often accessed frequently in a short period of time after it is written, the access frequency will become lower and lower as time goes on. If you can store files in the advanced pool for a short time and migrate them to the common pool after the access frequency is reduced, you can use lower-cost high-speed media to achieve higher storage performance, that is, file tiering. In actual use, by configuring the layering feature of the file, the file is written into the advanced pool and recorded in the index table. The MDS (metadata server) loads the index table and migrates files to...

AI Technical Summary

Problems solved by technology

If large files are migrated first, the water level of the advanced pool will drop faster, and more files can be stored. If small files are migrated first, the water level of the advanced pool will drop slowly, but in fact the files are stored out of order in the index table, lacking regularity , thus affecting the efficiency of file migration

[0004] There is currently no effective solution to the problem of low data file migration efficiency in the prior art

Images