Data reconstruction method and system wherein timing of data of data reconstruction is controlled in accordance with conditions when a failure occurs

• US 7,434,095 B2
• Filed: 09/06/2005
• Issued: 10/07/2008
• Est. Priority Date: 04/01/1991
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Kamo, Yoshihisa, Tanaka, Atsushi, Kakuta, Hitoshi
• Primary Examiner(s): Chu; Gabriel L
• Application Number:
  US11/218,610
• Publication Number:
  US 20060010343A1
• Time in Patent Office: 1,127 Days
• Field of Search: None
• US Class Current: 714/6.24
• CPC Class Codes: G06F 11/1084 Degraded mode, e.g. caused by single or multiple storage removals or disk failures
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G11C 29/74 using duplex memories, i.e. using dual copies

Abstract

A data storage system connectable to a host unit which issues data read/write requests to the data storage system, and which includes a plurality of disc units, and a controller connected to the disc units. A fault can occur in any of the disc units. The disc units store data in a plurality of data groups and error correcting data corresponding to each of the data groups. The controller, operable in different modes, performs processing of reconstructing data stored in any of the disc units in which a fault has occurred based on all other data belonging to any of the data groups to which the data to be reconstructed belongs and error correcting data corresponding to any of the data groups to which the data to be reconstructed belongs, and performs processing of data read/write requests from the host unit.

Claims

1. A storage system connectable to a computer comprising:

a plurality of disk units; and

a controller connected to the disk units;

wherein the disk units store data in a plurality of data groups and error correcting data corresponding to each of the data groups,

wherein the controller performs processing of reconstructing data stored in any of the disk units in which a fault has occurred wherein the controller performs processing of reconstructing data stored in any of the disk units in which a fault has occurred, producing faulty data, on all other data belonging to any based on all other data belonging to any of the data groups to which the data to be reconstructed belongs and error correcting data corresponding to any of the data groups to which the data to be reconstructed belongs, and performs processing of data read/write requests from the computer when a fault has occurred in any of the disk units, and

wherein the controller has a plurality of data reconstructing modes including a first mode wherein access or read/write requests from the computer is given preference when the time taken to reconstruct the faulty data is not more than a predetermined time limit, a second mode wherein the data reconstruction processing is given preference when the time taken to reconstruct the faulty data is more than the predetermined time limit, and a third mode wherein the data reconstruction processing is executed based on prepared programs.

2. The storage system according to claim 1, wherein said prepared programs are prepared corresponding to a plurality of conditions.

3. The storage system according to claim 2, wherein one of said conditions is job contents processed in the computer.

4. The storage system according to claim 2, wherein one of said condition is significance of the file stored in the disk units.

5. The storage system according to claim 2, when one of said conditions is changed, a suitable processing can be performed by replacing a prepared program corresponding to the one condition with another prepared program.

6. The storage system according to claim 1 wherein said first mode is a state wherein urgency of reconstruction is low, said second mode is a state wherein the urgency of the reconstruction is high, said third mode is a state wherein there is a condition between the urgency of the data reconstruction and the significance of the normal processing such as access or read/write.

7. The storage system according to claim 6, wherein the controller determines one of the data reconstructing modes to be performed based on a plurality of levels of urgency of data reconstruction.

8. The storage system according to claim 1, wherein the controller changes one of the data reconstructing modes which is performed by the storage system into another area of data reconstructing modes.

9. The storage system according to claim 1, wherein the controller comprises:

a faulty data reconstructing circuit; and

a faulty reconstruction control circuit,

wherein said faulty data reconstructing circuit detects any fault which has occurred in one of said plurality of disk units and informs said faulty reconstruction control circuit of the fault that has occurred, and

wherein said faulty reconstruction control circuit determines a condition of the fault including a determination whether a subsequent failure occurs when the data reconstruction of interest has not yet been completed and a determination whether a number of faulty discs for which data reconstruction is not completed exceeds a redundancy of the disk units so as to decide, and as to whether fault has occurred with respect to an access or read/write request from the computer in accordance with a result of the determination and informs a data loss or returns to a normal state by said faulty reconstruction control circuit.

10. The storage system according to claim 1, wherein the third mode is that the data reconstruction processing is made with changing the frequency in magnitude by determining whether or not the frequency of the processing of the normal access or read/write is more than the threshold.

11. The storage system according to claim 1, wherein the third mode is that the data reconstruction processing is made with setting the frequency of the data reconstruction processing or the amount of the data reconstruction within the unit time in accordance with the magnitude of the frequency of the processing of the normal access or read/write.