Capacity expansion method, system and device of metadata node cluster and storage medium

Abstract

The invention relates to a capacity expansion method and system of a metadata node cluster, equipment and a storage medium. A plurality of metadata services are deployed on each metadata node in the metadata node cluster; each metadata service independently corresponds to one metadata partition; the method comprises the following steps of: newly adding metadata node, determining at least one to-be-emigrated metadata partition needing to be emigrated in the original metadata node, and deploying a corresponding metadata service on the newly added metadata node, configuring the metadata service deployed on the newly added metadata node to be consistent with the configuration of the metadata service corresponding to the to-be-migrated metadata partition, and migrating the to-be-migrated metadata partition into the newly added metadata node. By adopting the method, the expansion complexity of the metadata node cluster can be reduced, and the expansion efficiency and effect of the metadata node cluster are improved.

Description

technical field [0001] The present application relates to the technical field of data storage, in particular to a method, system, device and storage medium for expanding the capacity of metadata node clusters. Background technique [0002] In cloud storage systems and distributed file systems, storage nodes are physical devices that store user data, metadata nodes are physical devices that store metadata, and metadata nodes are also responsible for managing data and operations of the entire system. Among them, metadata can be understood as data including information such as user data indexes, file names, and access permissions. [0003] With the continuous increase of user data, there are more and more metadata, and the pressure on metadata nodes is gradually increasing. In this case, it is necessary to expand the metadata node cluster formed by all metadata nodes in the system to improve the performance of the entire system. At present, commonly used expansion methods inc...

AI Technical Summary

Problems solved by technology

[0004] However, when using the above method to expand the capacity of the metadata node cluster, the metadata needs to be collected together for processing, and the original metadata needs to be split. data loss or corruption

In addition, when the amount of metadata that needs to be split is relatively large, the splitting process takes a long time, affecting system performance and normal operations of other businesses

Images