Storage control of storage media subject to write amplification effects

Patent metadata

Field	Value
Publication number	US-9996433-B2
Application number	US-201514685630-A
Country	US
Kind code	B2
Filing date	Apr 14, 2015
Priority date	Apr 29, 2014
Publication date	Jun 12, 2018
Grant date	Jun 12, 2018

Abstract

According to one aspect of the present invention, there is provided a method for performing storage control. Member storage media and a hot spare storage medium are identified in a storage system. The member storage media are members of a storage medium array, and the hot spare storage medium is for joining in the storage medium array when a member storage medium fails. Data on a member storage medium having a write amplification effect is migrated to the hot spare storage medium. In the member storage medium having a write amplification effect, an erase operation is performed on a storage medium where the migrated data is located. Embodiments of the present invention can alleviate adverse impact caused by a write amplification effect.

First claim

What is claimed is:

1. A method for performing storage control, comprising:
identifying member storage media and a hot spare storage medium in a storage system, wherein the member storage media are members of a storage medium array, and the hot spare storage medium is for joining in the storage medium array when a member storage medium fails;
migrating data on a member storage medium having a write amplification effect to the hot spare storage medium, wherein a variance of data volumes migrated from respective member storage media having a write amplification effect is smaller than a variance threshold; and
performing, in the member storage medium having a write amplification effect, an erase operation on a storage medium where the migrated data is located.

2. The method according to claim 1 , wherein cold data in member storage media having a write amplification effect is migrated to the hot spare storage medium in priority.

3. The method according to claim 1 , wherein migrating data on a member storage medium having a write amplification effect to the hot spare storage medium comprises:
identifying a data stride in the storage medium array, wherein the data stride is composed of data strips from respective member storage media, and mutual checks are enabled between the data strips;
determining a maximum number of data strips in the data stride that can be migrated to the hot spare storage medium based on a data redundancy of a structure of the storage medium array; and
migrating a number of data strips no more than the determined number to the hot spare storage medium.

4. The method according to claim 3 , further comprising:
determining, in response to detecting a storage medium failure, a lost data strip caused by the storage medium failure;
recovering the lost data strip with other data strips located in a same data stride as the lost data strip; and
storing the recovered data strip.

5. The method according to claim 4 , wherein storing the recovered data strip comprises:
storing the recovered data strip into a storage medium other than a storage medium where the data strips for recovering data are located.

6. The method according to claim 4 , further comprising:
in response to the hot spare storage medium storing data from member storage media of a plurality of storage medium arrays, reversely migrating data strips from storage medium arrays without a failing member storage medium to originating member storage media.

7. An apparatus for performing storage control, comprising:
an identifying module configured to identify member storage media and a hot spare storage medium in a storage system, wherein the member storage media are members of a storage medium array, and the hot spare storage medium is for joining in the storage medium array when a member storage medium fails;
a migrating module configured to migrate data on a member storage medium having a write amplification effect to the hot spare storage medium, wherein a variance of data volumes migrated from respective member storage media having a write amplification effect is smaller than a variance threshold; and
an erasing module configured to perform, in the member storage medium having a write amplification effect, an erase operation on a storage medium where the migrated data is located.

8. The apparatus according to claim 7 , wherein cold data in member storage media having a write amplification effect is migrated to the hot spare storage medium in priority.

9. The apparatus according to claim 7 , wherein the migrating module comprises:
a module configured to identify a data stride in the storage medium array, wherein the data stride is composed of data strips from respective member storage media, and mutual checks are enabled between the data strips;
a module configured to determine a maximum number of data strips in the data stride that can be migrated to the hot spare storage medium based on a data redundancy of a structure of the storage medium array; and
a module configured to migrate a number of data strips no more than the determined number to the hot spare storage medium.

10. The apparatus according to claim 9 , further comprising:
a determining module configured to determine, in response to detecting a storage medium failure, a lost data strip caused by the storage medium failure;
a recovering module configured to recover the lost data strip with other data strips located in a same data stride as the lost data strip; and
a storing module configured to store the recovered data strip.

11. The apparatus according to claim 10 , wherein the storing module comprises:
a module configured to store the recovered data strip into a storage medium other than a storage medium where the data strips for recovering data are located.

12. The apparatus according to claim 10 , further comprising:
a reverse migration module configured to, in response to the hot spare storage medium storing data from member storage media of a plurality of storage medium arrays, reversely migrate data strips from storage medium arrays without a failing member storage medium to originating member storage media.

13. A computer program product for performing storage control, the computer program product comprising:
one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions comprising:
program instructions to identify member storage media and a hot spare storage medium in a storage system, wherein the member storage media are members of a storage medium array, and the hot spare storage medium is for joining in the storage medium array when a member storage medium fails;
program instructions to migrate data on a member storage medium having a write amplification effect to the hot spare storage medium, wherein a variance of data volumes migrated from respective member storage media having a write amplification effect is smaller than a variance threshold; and
program instructions to perform, in the member storage medium having a write amplification effect, an erase operation on a storage medium where the migrated data is located.

14. The computer program product according to claim 13 , wherein cold data in member storage media having a write amplification effect is migrated to the hot spare storage medium in priority.

15. The computer program product according to claim 13 , wherein the program instructions to migrate data on a member storage medium having a write amplification effect to the hot spare storage medium comprise:
program instructions to identify a data stride in the storage medium array, wherein the data stride is composed of data strips from respective member storage media, and mutual checks are enabled between the data strips;
program instructions to determine a maximum number of data strips in the data stride that can be migrated to the hot spare storage medium based on a data redundancy of a structure of the storage medium array; and
program instructions to migrate a number of data strips no more than the determined number to the hot spare storage medium.

16. The computer program product according to claim 15 , wherein the program instructions stored on the one or more computer readable storage media further comprise:
program instructions to determine, in response to detecting a storage medium failure, a lost data strip caused by the storage medium failure;
program instructions to recover the lost data strip with other data strips located in a same data stride as the lost data strip; and
program instructions to store t

Assignees

• IBM

Inventors

• Shi yong hong
• Su Qian
• Sun Yu
• You Wei

Classifications

• G06F11/2094

  Redundant storage or storage space (G06F11/2056 takes precedence) · CPC title

• G06F11/2069

  Management of state, configuration or failover · CPC title

• G06F11/2038 Primary

  with a single idle spare processing component · CPC title

• G06F2211/1059

  Parity-single bit-RAID5, i.e. RAID 5 implementations · CPC title

• G06F11/1076 Primary

  Parity data used in redundant arrays of independent storages, e.g. in RAID systems · CPC title