Storage device management of unrecoverable logical block addresses for RAID data regeneration

• US 9,529,674 B2
• Filed: 06/18/2013
• Issued: 12/27/2016
• Est. Priority Date: 06/18/2013
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Pereira, David M., Nelogal, Chandrashekar
• Primary Examiner(s): Yi, David X
• Assistant Examiner(s): Perez-Velez, Rocio Del Mar
• Application Number:
  US13/920,731
• Publication Number:
  US 20140372697A1
• Time in Patent Office: 1,288 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/076 by exceeding a count or rate limit, e.g. word- or bit count limit
• CPC Class Codes: G06F 11/1088 Reconstruction on already foreseen single or plurality of spare disks

Abstract

A host processing system includes a processor, a RAID controller, and a data storage device coupled to the RAID controller and operable to detect first unrecoverable data at a first logical block address (LBA) of the data storage device, log the first LBA in an unrecoverable LBA table of the data storage device, provide the unrecoverable LBA table to the RAID controller, and in response to receiving a write to the first LBA, remove the first LBA from the unrecoverable LBA table.

Claims

1. A Redundant Array of Independent Drives (RAID) controller, comprising:

a memory; and

a processor operable to;

request a subset of an unrecoverable logical block address (LBA) table from a data storage device coupled to the RAID controller, wherein the subset of the unrecoverable LBA table includes unrecoverable LBAs associated with a particular physical block address;

determine, from the subset of the unrecoverable LBA table, a first LBA of the data storage device that includes first unrecoverable data;

reconstruct the first unrecoverable data at a time when the RAID controller is idle;

and issue a write of the reconstructed data to the first LBA;

determine, from the subset of the unrecoverable LBA table, a second LBA of the data storage device that includes second unrecoverable data;

determine that a number of LBAs that include unrecoverable data from the subset of the unrecoverable LBA table exceeds a first threshold failure rate based upon the determination that the second LBA includes second unrecoverable data; and

indicate that the storage device is in danger of failing based upon the determination that the number of LBAs that include unrecoverable data exceeds the first threshold failure rate.

2. The RAID controller of claim 1, wherein the subset of the unrecoverable LBA table comprises information related to a failure to read data stored at the first LBA.

3. The RAID controller of claim 2, wherein the information includes a source of detecting the failure.

4. The RAID controller of claim 3, wherein the source of detecting the failure includes at least one of a RAID controller read error, a data storage device self test error, and a data storage device background media scan.

5. The RAID controller of claim 2, wherein the information includes a physical block of the data storage device associated with the first LBA.

6. The RAID controller of claim 2, wherein the information includes a timestamp for the failure.

7. The RAID controller of claim 1, wherein the processor is further operable to:

determine, from the subset of the unrecoverable LBA table, a third LBA of the data storage device that includes third unrecoverable data;

determine that a rate at which the data storage device detected the first unrecoverable data and the third recoverable data exceeds a second threshold failure rate; and

indicate that the data storage device is in danger of failing based upon the determination that the rate at which the data storage device detected the first unrecoverable data and the third recoverable data exceeds the second threshold failure rate.

8. A host processing system, comprising:

a processor;

a Redundant Array of Independent Drives (RAID) controller; and

a data storage device coupled to the RAID controller and operable to;

detect first unrecoverable data at a first logical block address (LBA) of the data storage device;

log the first LBA in an unrecoverable LBA table of the data storage device;

detect second unrecoverable data at a second LBA of the data storage device;

log the second LBA in an unrecoverable LBA table of the data storage device;

provide a subset of the unrecoverable LBA table to the RAID controller; and

in response to receiving a write to the first LBA, remove the first LBA from the unrecoverable LBA table;

wherein the RAID controller is operable to;

request the subset of the unrecoverable LBA table from the data storage device, wherein the subset of the unrecoverable LBA table includes unrecoverable LBAs associated with a particular physical block address;

determine, from the subset of the unrecoverable LBA table, a third LBA of the data storage device that includes third unrecoverable data;

reconstruct the third unrecoverable data at a time when the RAID controller is idle; and

issue a write of the reconstructed data to the third LBA;

determine, from the subset of the unrecoverable LBA table, a fourth LBA of the data storage device that includes fourth unrecoverable data;

determine that a number of LBAs that include unrecoverable data from the subset of the unrecoverable LBA table exceeds a first threshold failure rate based upon the determination that the fourth LBA includes fourth unrecoverable data; and

indicate that the storage device is in danger of failing based upon the determination that the number of LBAs that include unrecoverable data exceeds the first threshold failure rate.

9. The host processing system of claim 8, wherein the log of the LBA in the unrecoverable LBA table comprises information related to a failure to read data stored at the first LBA.

10. The host processing system of claim 9, wherein the information includes a source of detecting the failure.

11. The host processing system of claim 10, wherein the source of detecting the failure includes at least one of a RAID controller read error, a self test error, and a background media scan.

12. The host processing system of claim 9, wherein the information includes a physical block associated with the first LBA.

13. The host processing system of claim 9, wherein the information includes a timestamp for the failure.

14. The host processing system of claim 8, wherein the data storage device is further operable to:

detect fifth unrecoverable data at a fifth LBA of the data storage device;

log the fifth LBA in the unrecoverable LBA table of the data storage device;

determine that a rate at which the data storage device detected the first unrecoverable data and the fifth recoverable data exceeds a second threshold failure rate; and

provide an indication to the RAID controller that the data storage device is in danger of failing based upon determining that the rate at which the data storage device detected the first unrecoverable data and the fifth recoverable data exceeds the second threshold failure rate.

15. The host processing system of claim 8, wherein:

the data storage device is further operable to;

detect fifth unrecoverable data at a fifth LBA of the data storage device;

log the fifth LBA in the unrecoverable LBA table of the data storage device; and

provide the unrecoverable LBA table to the RAID controller a second time; and

the RAID controller is further operable to;

determine that a rate at which the data storage device detected the first unrecoverable data and the fifth recoverable data exceeds a threshold failure rate; and

indicate that the data storage device is in danger of failing based upon determining that the rate at which the data storage device detected the first unrecoverable data and the fifth recoverable data exceeds the threshold failure rate.

16. A method comprising:

detecting, by a data storage device, first unrecoverable data at a first logical block address (LBA) of the data storage device;

logging the first LBA in an unrecoverable LBA table of the data storage device;

detecting second unrecoverable data at a second LBA of the data storage device;

logging the second LBA in the unrecoverable LBA table of the data storage device;

receiving, at a Redundant Array of Independent Drives (RAID) controller, a subset of the unrecoverable LBA table, wherein the subset of the unrecoverable LBA table includes unrecoverable LBAs associated with a particular source of failure;

determining, from the subset of the unrecoverable LBA table, a third LBA of the data storage device that includes third unrecoverable data;

reconstructing, at the RAID controller, the third unrecoverable data in response to receiving the subset of the unrecoverable LBA table, wherein the reconstructing is at a time when the RAID controller is idle;

issuing, by the RAID controller, a write of the reconstructed data to the third LBA; and

in response to receiving a write to the third LBA, removing the third LBA from the unrecoverable LBA table;

determining, from the subset of the unrecoverable LBA table, a fourth LBA of the data storage device that includes fourth unrecoverable data;

determining that a number of LBAs that include unrecoverable data from the subset of the unrecoverable LBA table exceeds a first threshold failure rate based upon the determination that the fourth LBA includes fourth unrecoverable data; and

indicating that the storage device is in danger of failing based upon the determination that the number of LBAs that include unrecoverable data exceeds the first threshold failure rate.

17. The method of claim 16, wherein the log of the first LBA in the unrecoverable LBA table comprises information related to a failure to read data stored at the first LBA.

18. The method of claim 17, wherein the information includes a source of detecting the failure, the source of detecting the failure including at least one of a RAID controller read error, a self test error, and a background media scan.

19. The method of claim 17, wherein the information includes at least one of a physical block associated with the first LBA, and a timestamp for the failure.

20. The method of claim 16, further comprising:

detecting fifth unrecoverable data at a fifth LBA of the data storage device;

logging the fifth LBA in the unrecoverable LBA table of the data storage device;

determining that a rate at which the data storage device detected the first unrecoverable data and the second fifth recoverable data exceeds a second threshold failure rate; and

indicating that the data storage device is in danger of failing based upon determining that the rate at which the data storage device detected the first unrecoverable data and the second fifth recoverable data exceeds the second threshold failure rate.