Server initiated predictive failure analysis for disk drives

• US 20040103337A1
• Filed: 06/20/2002
• Published: 05/27/2004
• Est. Priority Date: 06/20/2002
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Smith, Gordon James
• Granted Patent: US 7,631,067 B2
• Time in Patent Office: Days
• US Class Current: 714/2
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

The present invention provides a method, apparatus and program product for improving reliability in RAID/Server systems by monitoring the RAID/Server system at the system level for predictive failure analysis (PFA) triggering events. Examples of PFA triggering events include: rebuild operations, addition of new disk drives, a change in usage patterns, and suspected handling damage. Once a triggering event is detected, the RAID/Server system issues a command to the disk drives in the system to begin performing PFA. If a PFA error is detected on any of the drives, the error is reported to the RAID/Server system.

Claims

1. A method for performing predictive failure analysis (PFA) on one or more disk drives in a server system, the method comprising the steps of:

monitoring the server system at the system level for a PFA triggering event;

issuing a PFA initiation command to the one or more disk drives if a PFA triggering event is encountered on the server system;

performing PFA on the one or more disk drives; and

reporting a PFA error to the server system if any of the one or more disk drives fails the PFA.

2. The method of claim 1, wherein the PFA triggering event is a rebuild operation for the one or more disk drives.

3. The method of claim 1, wherein the PFA triggering event is the addition of new disk drives to the server system.

4. The method of claim 1, wherein the PFA triggering event is suspected handling damage to the one or more disk drives in the server system.

5. The method of claim 1, wherein the PFA triggering event is a change in usage patterns for the one or more disk drives in the server system.

6. The method of claim 1, wherein the method further includes the step of logging the results of the PFA in a reserved area of the one or more disk drives.

7. A computer system, comprising:

a server;

a Predictive Failure Analysis (PFA) monitor incorporated within the server, the PFA monitor capable of detecting a system level PFA triggering event, and issuing a PFA initiate command if a PFA triggering event is detected;

one or more Redundant Array of Independent Disk (RAID) units operably connected to the server; and

a plurality of disk drives incorporated within each of the one or more RAID units, wherein each of the plurality of disk drives is capable of receiving the PFA initiate command issued by the PFA monitor, the PFA command triggering a PFA measurement within each of the plurality of disk drives.

8. The computer system of claim 7, wherein the system level PFA triggering event is a rebuild operation for the plurality of disk drives.

9. The computer system of claim 7, wherein the system level PFA triggering event is the addition of new disk drives to the one or more RAID units.

10. The computer system of claim 7, wherein the system level PFA triggering event is suspected handling damage to the plurality of disk drives in the server system.

11. The computer system of claim 7, wherein the PFA triggering event is a change in usage patterns for the plurality of disk drives in the server system.

12. The computer system of claim 7, wherein each of the RAID units includes a RAID controller.

13. The computer system of claim 7, wherein the PFA command is routed from the PFA monitor of the server to the plurality of disks via the RAID controller.

14. A Redundant Array of Independent Disks (RAID), comprising:

a RAID controller;

a predictive failure analysis (PFA) monitor incorporated within the RAID controller, the PFA monitor capable of detecting a system level triggering event, and issuing a PFA initiate command if the PFA triggering event is detected; and

a plurality of disk drives incorporated within the RAID, wherein each of the plurality of disk drives is capable of receiving the PFA command issued by the PFA monitor, the PFA command triggering a PFA measurement within each of the plurality of disk drives.

15. The RAID of claim 14, wherein the system level PFA triggering event is a rebuild operation for the plurality of disk drives.

16. The RAID of claim 14, wherein the system level PFA triggering event is the addition of new disk drives to the RAID.

17. The RAID of claim 14, wherein the system level PFA triggering event is suspected handling damage to the one or more disk drives in the RAID.

18. The RAID of claim 14, wherein the PFA triggering event is a change in usage patterns for the one or more disk drives in the RAID.

19. A program product, comprising:

a predictive failure analysis (PFA) mechanism that monitors a server system at a system level for a PFA triggering event, issues a PFA initiation command to one or more disk drives if a PFA triggering event is encountered on the server system, performs PFA on the one or more disk drives, and reports a PFA error to the server system if any of the one or more disk drives fails the PFA, and

computer-readable signal bearing media bearing the PFA mechanism.

20. The program product of claim 19, wherein the computer-readable signal bearing media comprises recordable media.

21. The program product of claim 19, wherein the computer-readable signal bearing media comprises transmission media.