Deterministic preventive recovery from a predicted failure in a distributed storage system

• US 20050216800A1
• Filed: 01/21/2005
• Published: 09/29/2005
• Est. Priority Date: 03/24/2004
• Status: Active Grant

• Current Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Original Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Inventors: Bean, Robert George, Gaertner, Mark Allen, Burnett, Paul William, German, Robert John, Bicknell, Bruce Allen
• Granted Patent: US 7,526,684 B2
• Time in Patent Office: Days
• US Class Current: 714/718
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

A data storage subsystem in a distributed storage system having a plurality of predictive failure analyzing data storage devices. The subsystem furthermore has a circuit that is responsive to a predicted failure indication by a data storage device in relation to predetermined rules stored in memory for deterministically initiating a preventive recovery either by a data recovery procedure in the data storage device or by a fault tolerance storage arrangement in the subsystem.

Claims

1. A data storage subsystem in a distributed storage system, the subsystem having a plurality of predictive failure analyzing data storage devices and comprising a circuit that is responsive to a predicted failure indication by a data storage device in relation to predetermined rules stored in memory for deterministically initiating a preventive recovery by a data recovery procedure in the data storage device in cooperation with a fault tolerant storage arrangement in the subsystem.

2. The data storage subsystem of claim 1 wherein each data storage device comprises self-monitoring analysis and reporting technology.

3. The data storage subsystem of claim 1 wherein the subsystem employs a redundant array of independent drives.

4. The data storage subsystem of claim 3 wherein the rules are determinative in relation to whether the predicted failure involves data that is stored redundantly in the subsystem.

5. The data storage subsystem of claim 3 wherein the rules are determinative in relation to a defined criticality of the data associated with the predicted failure.

6. The data storage subsystem of claim 3 wherein the rules are determinative in relation to the extent of the predicted failure.

7. The data storage subsystem of claim 3 wherein the rules are determinative in relation to empirical performance information about the data storage device indicating the predicted failure.

8. A method for protecting stored data in a data storage subsystem having a plurality of data storage devices, comprising:

predicting a data transfer failure by one of the data storage devices; and

analyzing the predicted failure in relation to predetermined rules to deterministically initiate a preventive recovery routine by a data recovery procedure in the data storage device in cooperation with a fault tolerant storage arrangement in the subsystem.

9. The method of claim 8 wherein the predicting step is characterized by the data storage device comprising self-monitoring analysis and reporting technology.

10. The method of claim 8 wherein the analyzing step comprises recalling the preselected rules from a memory.

11. The method of claim 8 wherein the analyzing step is characterized by the rules relating whether the predicted failure involves data that is stored redundantly in the subsystem.

12. The method of claim 8 wherein the analyzing step is characterized by the rules relating an assigned criticality of data associated with the predicted failure.

13. The method of claim 8 wherein the analyzing step is characterized by the rules relating to an observed extent of the predicted failure.

14. The method of claim 8 wherein the analyzing step is characterized by the rules relating to empirical performance information about the data storage device.

15. A data storage subsystem comprising:

a plurality of predictive failure analyzing data storage devices defining a storage subsystem; and

means for recovering in accordance with predetermined rules stored in memory that deterministically minimize unscheduled maintenance of the subsystem.

16. The system of claim 15 wherein the means for recovering is characterized by selectively initiating a recovery either by a data recovery procedure in a data storage device of the plurality indicating a predicted failure or by a fault tolerance data storage arrangement in the subsystem.

17. The system of claim 15 wherein the means for recovering is characterized by determining whether a predicted failure involves data that is stored redundantly in the subsystem.

18. The system of claim 15 wherein the means for recovering is characterized by determining an assigned criticality of data associated with a predicted failure.

19. The system of claim 15 wherein the means for recovering is characterized by determining an observed extent of a predicted failure.

20. The system of claim 15 wherein the means for recovering is characterized by determining empirical performance information about a data storage device predicting a failure.