Method, device and computer program products for storage management

Patent metadata

Field	Value
Publication number	US-11829604-B2
Application number	US-202017020033-A
Country	US
Kind code	B2
Filing date	Sep 14, 2020
Priority date	Mar 12, 2020
Publication date	Nov 28, 2023
Grant date	Nov 28, 2023

Abstract

Techniques for storage management involve determining, in response to a detection that a disk changes from an unavailable state to an available state, at least one candidate storage unit associated with the disk; acquiring historical access information about the at least one candidate storage unit, the historical information comprising information related to a write request directed to the at least one candidate storage unit when the disk is in the unavailable state; determining a target storage unit from the at least one candidate storage unit based on the historical access information; and rebuilding the target storage unit. Such techniques may, in a low-cost manner, improve rebuilding efficiency and reliability of a storage system.

First claim

The invention claimed is:

1. A method for storage management, comprising:
determining, in response to a detection that a disk changes from an unavailable state to an available state, a plurality of candidate RAID stripes associated with the disk, wherein each one of the candidate RAID stripes is built using a plurality of disk slices, wherein each one of the candidate RAID stripes is built at least in part using a respective unique one of a plurality of disk slices located on the disk, and wherein determining the plurality of candidate RAID stripes associated with the disk comprises identifying all RAID stripes built using any one of the plurality of disk slices located on the disk as the plurality of candidate RAID stripes;
requesting, by a first storage node from a second storage node, historical access information about the candidate RAID stripes, wherein the first storage node and the second storage node are peer storage nodes, wherein the candidate RAID stripes are accessible through both the first storage node and the second storage node, the historical information comprising information related to a write request directed to the candidate RAID stripes when the disk is in the unavailable state;
prior to receiving the historical access information about the candidate RAID stripes from the second storage node, in response to detecting that the second storage node is offline, setting, by the first storage node, access flags corresponding to all of the candidate RAID stripes, wherein the access flags corresponding to the candidate RAID stripes indicate whether the corresponding candidate RAID stripes were written while the disk was in the unavailable state, and wherein the set access flags corresponding to all of the candidate RAID stripes indicate that all of the candidate RAID stripes were written while the disk was in the unavailable state;
determining a target RAID stripe from the candidate RAID stripes based on the historical access information, wherein determining the target RAID stripe from the candidate RAID stripes is performed based at least in part on the access flags corresponding to all of the candidate RAID stripes; and
rebuilding the target RAID stripe.

2. The method of claim 1 , the method further comprising:
removing, for candidate RAID stripes other than the target RAID stripe in the plurality of candidate RAID stripes, disabling of a write operation when the disk is in the unavailable state.

3. The method of claim 1 , further comprising: acquiring, from the write request received when the disk is in the unavailable state, an identifier of a RAID stripe to which the write request is directed; and updating, according to a determination that the identifier corresponds to one of the candidate RAID stripes, an access flag corresponding to that one of the candidate RAID stripes, for indicating that the write request directed to that one of the candidate RAID stripes exists when the disk is in the unavailable state.

4. The method of claim 1 , further comprising:
removing, for the target RAID stripe and in response to completion of the rebuilding, disabling of a write operation when the disk is in the unavailable state.

5. The method of claim 1 , further comprising:
clearing, in response to completion of the rebuilding, an access flag directed to the target RAID stripe, the access flag indicating that a write request directed to the target RAID stripe exists when the disk is in the unavailable state.

6. An electronic device, comprising:
a processor; and
a memory coupled to the processor and storing instructions to be executed, wherein when executed by the processor, the instructions cause the electronic device to perform actions comprising:
determining, in response to a detection that a disk changes from an unavailable state to an available state, a plurality of candidate RAID stripes associated with the disk, wherein each one of the candidate RAID stripes is built using a plurality of disk slices, wherein each one of the candidate RAID stripes is built at least in part using a respective unique one of a plurality of disk slices located on the disk, and wherein determining the plurality of candidate RAID stripes associated with the disk comprises identifying all RAID stripes built using any one of the plurality of disk slices located on the disk as the plurality of candidate RAID stripes;
requesting, by a first storage node from a second storage node, historical access information about the candidate RAID stripes, wherein the first storage node and the second storage node are peer storage nodes, wherein the candidate RAID stripes are accessible through both the first storage node and the second storage node, the historical information comprising information related to a write request directed to the candidate RAID stripes when the disk is in the unavailable state;
prior to receiving the historical access information about the candidate RAID stripes from the second storage node, in response to detecting that the second storage node is offline, setting, by the first storage node, access flags corresponding to all of the candidate RAID stripes, wherein the access flags corresponding to the candidate RAID stripes indicate whether the corresponding candidate RAID stripes were written while the disk was in the unavailable state, and wherein the set access flags corresponding to all of the candidate RAID stripes indicate that all of the candidate RAID stripes were written while the disk was in the unavailable state;
determining a target RAID stripe from the candidate RAID stripes based on the historical access information; and
rebuilding the target storage unit.

7. The device of claim 6 , the actions further comprising:
removing, for candidate RAID stripes other than the target RAID stripe in the plurality of candidate RAID stripes, disabling of a write operation when the disk is in the unavailable state.

8. The device of claim 6 , wherein the actions further comprise: acquiring, from the write request received when the disk is in the unavailable state, an identifier of a RAID stripe to which the write request is directed; and updating, according to a determination that the identifier corresponds to one of the candidate RAID stripes, an access flag corresponding to that one of the candidate RAID stripes, for indicating that the write request directed to that one of the candidate RAID stripes exists when the disk is in the unavailable state.

9. The device of claim 6 , wherein the actions further comprise:
clearing, in response to completion of the rebuilding, an access flag directed to the target RAID stripe, the access flag indicating that a write request directed to the target RAID stripe exists when the disk is in the unavailable state.

10. A computer program product having a non-transitory computer readable medium which stores a set of instructions to perform storage management; the set of instructions, when carried out by computerized circuitry, causing the computerized circuitry to perform a method of:
determining, in response to a detection that a disk changes from an unavailable state to an available state, a plurality of candidate RAID stripes associated with the disk, wherein each one of the candidate RAID stripes is built using a plurality of disk slices, wherein each one of the candidate RAID stripes is built at least in part using a respective unique one of a plurality of disk slices located on the disk, and wherein determining the plurality of candidate RAID stripes associated with the disk comprises identifying all RAID stripes built using any one of the plurality of disk slices located on the disk as the plurality of candidate RAID stripes;
requesting, by a first storage node from a seco

Assignees

• Emc Ip Holding Co Llc

Inventors

• Gao Hongpo
• Xu Xinlei
• Yang Lifeng
• Kang Jianbin
• Han Geng
• Zhao Zhenhua

Classifications

• G06F3/0619 Primary

  in relation to data integrity, e.g. data losses, bit errors · CPC title

• G06F3/0622

  in relation to access · CPC title

• G06F3/0653

  Monitoring storage devices or systems · CPC title

• G06F3/0659

  Command handling arrangements, e.g. command buffers, queues, command scheduling · CPC title

• G06F3/0683

  Plurality of storage devices · CPC title