Storage system having RAID level changing function

• US 7,827,351 B2
• Filed: 01/09/2008
• Issued: 11/02/2010
• Est. Priority Date: 07/31/2007
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Sato, Takao, Okami, Yoshinori, Suetsugu, Michio
• Primary Examiner(s): LANE, JOHN A
• Application Number:
  US11/971,500
• Publication Number:
  US 20090037656A1
• Time in Patent Office: 1,028 Days
• Field of Search: 711/114, 711/165, 711/170
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/1096 Parity calculation or recalculation after configuration or reconfiguration of the system
• CPC Class Codes: G06F 2211/1057 Parity-multiple bits-RAID6, i.e. RAID 6 implementations
• CPC Class Codes: G06F 2211/1059 Parity-single bit-RAID5, i.e. RAID 5 implementations

Abstract

A plurality of storage devices, provided in a storage system, comprise not less than two member storage devices, which are storage devices provided as members of a RAID group, and a spare storage device, which is not a member of the RAID group. A controller, provided in the storage system, uses the spare storage device to carry out the changing of the RAID level of the RAID group from a first RAID level to a second RAID level.

Claims

1. A storage system, comprising:

a plurality of storage devices having not less than two member storage devices which are storage devices provided as members of a RAID group, and either one or a plurality of spare storage devices which are spare storage devices that are not members of the RAID group; and

a controller for changing a RAID level of the RAID group from a first RAID level to a second RAID level in use of a spare storage device selected from among the either one or a plurality of spare storage devices,

wherein the second RAID level has higher fault tolerance than the first RAID level, and

wherein the controller changes the RAID level of the RAID group from the first RAID level to the second RAID level in use of the selected spare storage device when an error rate of a certain member storage device of the not less than two member storage devices exceeds a first threshold.

2. The storage system according to claim 1, wherein

the controller, when the error rate of a certain member storage device of the not less than two member storage devices exceeds a second threshold, carries out an information backup process in which an information element stored in this certain member storage device is written to a separate spare storage device selected from among the plurality of spare storage devices, and

wherein the first threshold is lower than the second threshold.

3. The storage system according to claim 2, wherein the controller returns the RAID level of the RAID group from the second RAID level to the first RAID level subsequent to the information backup processing.

4. The storage system according to claim 3, wherein

the controller returns the RAID level of the RAID group from the second RAID level to the first RAID level when the error rate of a member storage device other than the certain member storage device is not more than a prescribed rate, and/or when the error rates of the respective member storage devices are not more than a prescribed rate after passage of a fixed time period.

5. The storage system according to claim 1, wherein

the second RAID level is a RAID level in which at least one parity is created for each stripe which is a storage area constituting the RAID group, and

wherein the controller writes an information element stored in the certain member storage device to the selected spare storage device, and writes the parity of each stripe to the certain member storage device.

6. The storage system according to claim 5, wherein the controller returns the RAID level of the RAID group from the second RAID level to the first RAID level by removing the certain member storage device from the RAID group, and managing the selected spare storage device as a member of the RAID group.

7. The storage system according to claim 1, wherein

the second RAID level is a RAID level in which at least one parity is created for each stripe which is a storage area constituting the RAID group,

wherein the selected spare storage device is provided in plurality, and

the controller distributes a plurality of parities, for a plurality of stripes constituting the RAID group, to the selected plurality of spare storage devices.

8. The storage system according to claim 1, wherein

the second RAID level is a RAID level in which at least one parity is created for each stripe which is a storage area constituting the RAID group, and

wherein the controller distributes a plurality of parities, for a plurality of stripes constituting the RAID group, to the selected spare storage device and the not less than two member storage devices.

9. The storage system according to claim 1, wherein changing from the first RAID level to the second RAID level requires reading out information elements from the not less than two member storage devices, for each stripe which is a storage area constituting the RAID group,

a logical volume is formed on based a storage space of the RAID group, and the logical volume is constituted from a plurality of pages, which are a plurality of storage areas, and

the controller is constituted such that, when a write command received from outside the storage system specifies a virtual storage area in a virtual volume, the controller allocates a selected page from the plurality of pages to the virtual storage area, and writes an information element which conforms to the write command to the allocated page, and in changing from the first RAID level to the second RAID level, the controller reads out an information element only for the stripe corresponding to the page allocated to the virtual logical volume.

10. The storage system according to claim 1, wherein the either one or a plurality of spare storage devices is shared for use with a plurality of RAID groups.

11. The storage system according to claim 1, wherein the controller has a cache memory, and is constituted so as to receive a write command from outside the storage system, write an information element based on this write command to the cache memory, and, in this case, report write-completion to an outside unit, and, asynchronously thereto, write this information element to the RAID group, and if the RAID level of the RAID group is in a process of being changed, the controller holds off on writing the information element stored in the cache memory to the RAID group, and writes this information element to the RAID group subsequent to the completion of the RAID level change.

12. The storage system according to claim 11, wherein

the controller comprises a higher-level I/F, which is an interface device for the outside unit, and a lower-level I/F, which is an interface device for the plurality of storage devices,

wherein the higher-level I/F is constituted so as to carry out reception of a write command from the outside unit, writing of an information element based on this write command to the cache memory, and reporting of write-completion to the outside unit, and

wherein the lower-level I/F is constituted so as to carry out writing of the information element stored in the cache memory to the RAID group, change of the RAID level of the RAID group, holding off of writing the information element stored in the cache memory to the RAID group if the RAID level is in a process of being changed, and writing of this information element to the RAID group subsequent to completion of the RAID level change.

13. The storage system according to claim 1, wherein the second RAID level is a level in which many more parities are created for each stripe which is a storage area constituting the RAID group than in the first RAID level.

14. The storage system according to claim 13, wherein the controller, when an error rate of a certain member storage device of the not less than two member storage devices exceeds a first threshold, executes, for each stripe constituting the RAID group, read-out of not less than two information elements stored in a stripe, creation of a parity based on the not less than two information elements, and write of the created parity to the selected spare storage device, and the controller changes the RAID level of the RAID group from the first RAID level to the second RAID level by managing the selected spare storage device as another member of the RAID group.

15. The storage system according to claim 14, wherein the controller is constituted such that, when the error rate of a certain member storage device of the not less than two member storage devices exceeds a second threshold, the controller carries out an information backup process for writing an information element stored in this certain member storage device to a separate spare storage device selected from the plurality of spare storage devices, and

wherein the first threshold is lower than the second threshold.

16. The storage system according to claim 15, wherein subsequent to the information backup process, the controller returns the RAID level of the RAID group from the second RAID level to the first RAID level by writing the information element stored in the selected separate spare storage device to a post-replacement member storage device, which is the storage device mounted in place of the certain member storage device subsequent to the certain member storage device being removed from the storage system, and by removing the selected spare storage device from the RAID group members.

17. The storage system according to claim 14, wherein the controller returns the RAID level of the RAID group from the second RAID level to the first RAID level by writing the information element stored in the certain member storage device to the selected spare storage device for each stripe constituting the RAID group, writing the parity of each the stripe to the certain member storage device, and removing the certain member storage device from the RAID group.

18. A storage system comprising:

a plurality of storage devices having not less than two member storage devices which are storage devices provided as members of a RAID group, and a spare storage device which is not a member of this RAID group; and

a controller for carrying out an information backup process, which writes an information element stored in a certain member storage device to the spare storage device when an error rate of this certain member storage device of the not less than two member storage devices exceeds a second threshold,

wherein the controller changes a RAID level of the RAID group from a first RAID level to a second RAID level, which features higher fault tolerance than the first RAID level, prior to carrying out the information backup process.