Capacity expansion method and system of storage cluster and related device

Abstract

The invention provides a capacity expansion method of a storage cluster, which comprises the following steps: receiving a capacity expansion instruction, and determining a new capacity expansion nodecorresponding to the capacity expansion instruction; traversing all the placement groups, and determining the reference count of the OSD corresponding to each placement group, wherein the reference count of each OSD is the same as the number of the placement groups borne by the OSD; setting the OSD with the maximum reference count in each placement group as a to-be-determined OSD; and calling a data distribution interface to select an optimal OSD from the new expansion nodes to replace the OSD to be determined. According to the invention, the OSD with the maximum reference technology in each placement group is only replaced by the OSD in the new expansion node, and data migration between the interiors of the old nodes is avoided, so that the expansion reconstruction efficiency is improved,the resource utilization rate of the system during expansion is reduced, and the performance of the storage system is improved. The invention further provides a capacity expansion system of the storage cluster, a computer readable storage medium and storage equipment, which have the above beneficial effects.

Description

technical field [0001] The present application relates to the field of data storage, in particular to a storage cluster expansion method, system and related devices. Background technique [0002] In a distributed storage system, the stored content is cut according to a fixed size. This piece of fixed-size data is called an object, and PG (Placement Group, a carrier for placing objects) is an aggregation of multiple objects. The correspondence between each PG and OSD (Object-based Storage Device, object storage device) is through the data distribution algorithm. One PG corresponds to multiple OSDs. OSD can be simply understood as a physical disk, and one OSD can There are multiple PGs, and the stored content is placed on disk. When performing node expansion, the data distribution algorithm will re-select OSDs for all PGs through a series of calculations. After the selection is completed, data migration is required for the newly added OSD members in the PG, but it is easy to ...

AI Technical Summary

Problems solved by technology

When performing node expansion, the data distribution algorithm will re-select OSDs for all PGs through a series of calculations. After the selection is completed, data migration is required for the newly added OSD members in the PG, but it is easy to exist during the data migration process. After a PG reselects members, the newly added members are still the OSD of the old node, but because there is no data of the PG on the OSD, data migration is also required, so that there will be additional migration during capacity expansion, which not only occupies system resources , while reducing the expansion efficiency

Images