Method and device for data reconstruction

Abstract

The invention discloses a method and a device for data reconstruction. The method comprises the following steps: determining the number T of hot standby VDs (Virtual Disks), wherein the hot standby VDs are those which do not participate in VDG (Virtual Disk Group) reconstruction on a normal disk; determining the number S of the hot standby VDs to be converted into idle VDs from the T hot standby VDs, wherein T and S are both natural integers which are greater than 0, and S is smaller than T; selecting the S hot standby VDs from the T hot standby VDs, converting the S hot standby VDs into the idle VDs, and converting the S idle VDs on a dilatation disk into hot standby VDs; enabling the hot standby VDs on the normal disk and the dilatation disk to participate in athe VDG reconstruction. By applying the method disclosed by the invention, the dilatation disk which is newly added is enabled to more participate in a reconstruction process, so that the efficiency of data reconstruction is increased, and meanwhile, the equalization time after reconstruction can be shortened.

Description

technical field [0001] The present application relates to the technical field of data storage, and in particular to a data reconstruction method and device. Background technique [0002] With the rapid development of science and technology and the widespread application of computer technology, people have higher and higher requirements on the performance of storage devices. As a mature technology, RAID (Redundant Arrays of Independent Disks, Redundant Array of Independent Disks) technology has been widely used in various occasions of data storage. RAID technology uses multiple PDs (Physical Disks) to build a storage pool. Each PD is divided into multiple VDs (Virtual Disks). Group, virtual disk group), providing logical resources to upper-layer clients for reading and writing. [0003] Under the block virtualization structure of the storage pool, when a certain PD fails, in order to prevent other PDs from also failing and causing the upper-layer business to be affected, an...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, during the VDG rebuilding process, if enough hot-standby VDs can be applied for on the original PD, the new PDs that have been expanded will seldom participate in the rebuilding process, and the number of VDGs allowed to be rebuilt at the same time is small , which leads to a long time spent in the reconstruction process. In addition, after the reconstruction is completed, the utilization of all PDs (including the original PD and the new PD that has been expanded) can be balanced. Because the new PD is in the reconstruction process Basically, no read and write operations are involved in any read and write operations, and the utilization rate of the new PD after reconstruction is quite different from that of the original PD, resulting in a longer equalization time.

Images