Efficient system for predicting and processing storage subsystem failure

• US 5,832,204 A
• Filed: 06/23/1997
• Issued: 11/03/1998
• Est. Priority Date: 03/12/1996
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Apperley, Norman, Brady, James Thomas, Gerdt, Steven, Johnson, Alden B.
• Primary Examiner(s): Hua, Ly
• Application Number:
  US08/880,610
• Time in Patent Office: 498 Days
• Field of Search: 395/182.04, 395/183.18, 395/182.14, 395/182.03, 395/182.05, 395/183.16, 371/40.11, 371/51.1, 360/77.03, 360/77.08, 360/51, 360/75
• US Class Current: 714/42
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/1088 Reconstruction on already foreseen single or plurality of spare disks
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence

Abstract

Predictive failure analysis of a storage subsystem is efficiently conducted and data quickly recovered from a failed Read operation. This may be implemented in a storage system including a host coupled to a supervising processor that couples to a parity-equipped RAID storage subsystem having multiple HDAs each including an HDA controller and at least one storage medium. In one embodiment, when an HDA experiences an error during a Read attempt, the HDA transmits a recovery alert signal to the supervising processor; then, the processor and HDA begin remote and local recovery processes in parallel. The first process to complete provides the data to the host, and the second process is aborted. In another embodiment, an HDA'"'"'s PFA operations are restricted to idle times of the HDA. A different embodiment limits HDA performance of PFA to times when the processor is conducting data reconstruction. Another embodiment monitors HDA errors at the supervisory processor level, initiating an HDA'"'"'s PFA operations when errors at that HDA have a certain characteristic, such as a predetermined frequency of occurrence.

Claims

1. A method for data recovery in a storage system including a host coupled to a supervising processor coupled to a parity-equipped RAID storage subsystem having multiple head disk assemblies ("HDA") each HDA including an HDA controller and at least one storage medium, said method comprising the steps of:

the supervising processor receiving from said host a Read request for reading target data;

the supervising processor directing a first HDA to read the target data;

the first HDA attempting to read the target data and detecting a data error during the attempt;

the supervising processor executing a recovery process to reconstruct the target data by combining supplementary data and stored parity, said supplementary data comprising data corresponding to the target data and stored elsewhere in the RAID storage system than the first HDA, and said stored parity comprising parity corresponding to the target data and supplementary data and stored in the RAID storage subsystem;

concurrently with the supervising processor executing the recovery process, the first HDA performing a selected predictive failure analysis ("PFA") to predict future failure of the at least one storage medium of the first HDA; and

after completion of the recovery process, providing reconstructed target data from the supervising processor to the host.

2. The method of claim 1, the at least one storage medium being a single storage medium.

3. The method of claim 1, the at least one storage medium comprising multiple storage media.

4. The method of claim 1, further comprising the steps of:

determining whether the recovery process or PFA completes first; and

aborting the PFA in the event the recovery process completes first.

5. The method of claim 1, the step of the first HDA performing a selected PFA further including the steps of identifying potential causes of the predicted future failure.

6. A data storage medium tangibly embodying a computer-processor-readable program of instructions for causing a supervising processor and a first one of a multiple of head disk assemblies to perform method steps for data recovery in a storage system, which storage system including a parity-equipped RAID storage subsystem coupled to said supervising processor and having said multiple of head disk assemblies ("HDA"), each HDA including an HDA controller and at least one storage medium, said method steps comprising:

the supervising processor receiving from said host a Read request for reading target data;

the supervising processor directing a first HDA to read the target data;

the first HDA attempting to read the target data and detecting a data error during the attempt;

the supervising processor executing a recovery process to reconstruct the target data by combining supplementary data and stored parity, said supplementary data comprising data corresponding to the target data and stored elsewhere in the RAID storage system than the first HDA, and said stored parity comprising parity corresponding to the target data and supplementary data and stored in the RAID storage subsystem;

concurrently with the supervising processor executing the recovery process, the first HDA performing a selected predictive failure analysis ("PFA") to predict future failure of the at least one storage medium of the first HDA; and

after completion of the recovery process, providing reconstructed target data from the supervising processor to the host.

7. The data storage medium of claim 6, the at least one storage medium being a single storage medium.

8. The data storage medium of claim 6, the at least one storage medium comprising multiple storage media.

9. The data storage device of claim 6, further comprising the steps of:

determining whether the recovery process or PFA completes first; and

aborting the PFA in the event the recovery process completes first.

10. The data storage medium of claim 6, the step of the first HDA performing a selected PFA further including the steps of identifying potential causes of the predicted future failure.