Virtual failure domains for storage systems

Patent metadata

Field	Value
Publication number	US-10114716-B2
Application number	US-201514947642-A
Country	US
Kind code	B2
Filing date	Nov 20, 2015
Priority date	Nov 20, 2015
Publication date	Oct 30, 2018
Grant date	Oct 30, 2018

Abstract

A method for storage systems improvement includes collecting information that indicates one or more failure correlations for disks in a storage system. The disks are then separated into a plurality of virtual failure domains based on the indicated one or more failure correlations. The method then determines that all data objects of a set of redundant data objects are included in a first virtual failure domain. Responsive to determining that all data objects of the set of redundant data objects are included in the first virtual failure domain, the method then migrates at least one data object of the set of redundant data objects from a first disk in the first virtual failure domain to a second disk in a second virtual failure domain.

First claim

What is claimed is:

1. A method comprising:
collecting, by one or more processors, information that indicates one or more failure correlations for disks in a storage system;
defining, by one or more processors, each disk as a vector of parameters associated with the respective disk, wherein the parameters are based on the information that indicates one or more failure correlations, and wherein the parameters for each respective disk include the respective disk's physical location inside the storage system, the respective disk's manufacture data, and the respective disk's performance/usage parameters, wherein the performance/usage parameters include a number of head crashes and a number of bad sectors;
separating, by one or more processors, the disks into a plurality of virtual failure domains based on the parameters of their corresponding vectors;
determining, by one or more processors, that all data objects of a set of redundant data objects are included in a first virtual failure domain; and
responsive to determining that all data objects of the set of redundant data objects are included in the first virtual failure domain, migrating, by one or more processors, at least one data object of the set of redundant data objects from a first disk in the first virtual failure domain to a second disk in a second virtual failure domain.

2. The method of claim 1 , wherein the manufacture data includes the following:
a manufacturer name;
a model number;
a manufacturer batch number; and
a serial number.

3. The method of claim 1 , wherein the physical location of a respective disk inside the storage system includes a position of the respective disk in a rack; and
the performance/usage parameters of a respective disk further include a disk age.

4. The method of claim 1 , wherein the performance/usage parameters of a respective disk further include the following:
a number of read/writes for the disk;
an operating temperature; and
a number of read/write errors.

5. The method of claim 4 , wherein the performance/usage parameters of a respective disk further include Self-Monitoring, Analysis, and Reporting Technology (SMART) parameters.

6. The method of claim 1 , wherein the virtual failure domains are determined during a system runtime by using machine learning (ML) classification algorithms configured to identify groups based on failure ratio correlations.

7. The method of claim 1 , further comprising:
discovering, by one or more processors, one or more sub-groups of the plurality of virtual failure domains during runtime; and
separating, by one or more processors, one or more of the disks into the one or more sub-groups.

8. A computer program product for storage systems improvement, 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 collect information that indicates one or more failure correlations for disks in a storage system;
program instructions to define each disk as a vector of parameters associated with the respective disk, wherein the parameters are based on the information that indicates one or more failure correlations, and wherein the parameters for each respective disk include the respective disk's physical location inside the storage system, the disk's manufacture data, and the respective disk's performance/usage parameters, wherein the performance/usage parameters include a number of head crashes and a number of bad sectors;
program instructions to separate the disks into a plurality of virtual failure domains based on the parameters of their corresponding vectors;
program instructions to determine that all data objects of a set of redundant data objects are included in a first virtual failure domain; and
program instructions to, responsive to determining that all data objects of the set of redundant data objects are included in the first virtual failure domain, migrate at least one data object of the set of redundant data objects from a first disk in the first virtual failure domain to a second disk in a second virtual failure domain.

9. The computer program product of claim 8 , wherein the manufacture data includes the following:
a manufacturer name;
a model number;
a manufacturer batch number; and
a serial number.

10. The computer program product of claim 8 , wherein:
the physical location of a respective disk inside the storage system includes a position of the respective disk in a rack; and
the performance/usage parameters of a respective disk further include a disk age.

11. The computer program product of claim 8 , wherein the performance/usage parameters of a respective disk further include the following:
a number of read/writes for the disk;
an operating temperature; and
a number of read/write errors.

12. The computer program product of claim 11 , wherein the performance/usage parameters of a respective disk further include Self-Monitoring, Analysis, and Reporting Technology (SMART) parameters.

13. The computer program product of claim 8 , wherein the virtual failure domains are determined during a system runtime by using machine learning (ML) classification algorithms configured to identify groups based on failure ratio correlations.

14. The computer program product of claim 8 , further comprising:
program instructions to discover one or more sub-groups of the plurality of virtual failure domains during runtime; and
program instructions to separate one or more of the disks into the one or more sub-groups.

15. A computer system for storage systems improvement, the computer system comprising:
one or more computer processors;
one or more computer readable storage media;
program instructions stored on the computer readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to collect information that indicates one or more failure correlations for disks in a storage system;
program instructions to define each disk as a vector of parameters associated with the respective disk, wherein the parameters are based on the information that indicates one or more failure correlations, and wherein the parameters for each respective disk include the respective disk's physical location inside the storage system, the respective disk's manufacture data, and the respective disk's performance/usage parameters, wherein the performance/usage parameters include a number of head crashes and a number of bad sectors;
program instructions to separate the disks into a plurality of virtual failure domains based on the parameters of their corresponding vectors;
program instructions to determine that all data objects of a set of redundant data objects are included in a first virtual failure domain; and
program instructions to, responsive to determining that all data objects of the set of redundant data objects are included in the first virtual failure domain, migrate at least one data object of the set of redundant data objects from a first disk in the first virtual failure domain to a second disk in a second virtual failure domain.

16. The computer system of claim 15 , wherein the manufacture data includes the following:
a manufacturer name;
a model number;
a manufacturer batch number; and
a serial number.

17. The computer system of claim 15 , wherein the performance/usage parameters of a respective disk further include the following:
a number of read/writes for the disk;

Assignees

• IBM

Inventors

• Epstein Amir
• Factor Michael E
• Harnik Danny
• Kat Ronen I
• Kolodner Elliot K
• Sotnikov Dmitry

Classifications

• G06F11/00

  Error detection; Error correction; Monitoring (error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00) · CPC title

• G06F3/0619

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

• G06F2201/805

  Real-time · CPC title

• G06F11/2097 Primary

  maintaining the standby controller/processing unit updated (initialisation or re-synchronisation thereof G06F11/1658 and subgroups) · CPC title

• G06F3/0683

  Plurality of storage devices · CPC title