System and method for reorganizing data in a raid storage system

• US 20050166085A1
• Filed: 02/28/2005
• Published: 07/28/2005
• Est. Priority Date: 06/24/2002
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Mark J. Thompson, Stephen M. Schultz
• Inventors: Thompson, Mark J., Schultz, Stephen M.
• Granted Patent: US 7,281,089 B2
• Time in Patent Office: Days
• US Class Current: 714/6
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 2211/1035 Keeping track, i.e. keeping track of data and parity changes
• CPC Class Codes: G06F 3/0605 by facilitating the interaction with a user or administrator
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G11B 2220/2516 Hard disks
• CPC Class Codes: G11B 2220/415 Redundant array of inexpensive disks [RAID] systems
• CPC Class Codes: G11B 27/002 Programmed access in sequence to a plurality of record carriers or indexed parts, e.g. tracks, thereof, e.g. for editing; transfer of record carriers from magazine G11B15/68, G11B17/10; G11B17/22
• CPC Class Codes: G11B 27/105 of operating discs

Abstract

Storage controllers and related methods of operation are described. One method tracks storage access operations to a logical disk. The method persistently identifies storage blocks that have been accessed. Responsive to the identifying, the method selectively skips subsequent operations to storage blocks that have not been accessed.

Claims

1. A method of operating a storage controller comprising:

receiving data access requests containing a logical block address and a count of blocks involved in the access request;

persistently storing a record indicating whether particular logical block addresses have been accessed since creation;

in response to detecting a condition that requires an operation that accesses a logical block address range, consulting the persistently stored record as to whether the logical block address range has been previously accessed; and

,

skipping the operation when the logical block address range has not been previously accessed.

2. The method of claim 1, wherein the persistent record is stored in one or more reserved sectors of at least one disk drive in an array of disks.

3. The method of claim 1, wherein the persistent record is stored in non-volatile memory of the controller.

4. The method of claim 1, wherein the operation comprises a rebuild operation for a selected LBA (logical block address) range.

5. The method of claim 1, wherein the operation comprises a capacity expansion operation for a selected LBA (logical block address) range.

6. The method of claim 1, wherein the operation comprises a data migration operation for a selected LBA (logical block address) range.

7. The method of claim 1, wherein the act of skipping can be selectively disabled.

8. The method of claim 1, wherein the record indicates whether any of a particular range of block addresses has been previously written.

9. The method of claim 1, wherein the record indicates whether any of a particular range of block addresses has been previously read.

10. The method of claim 1 further comprising:

defining a plurality of zones, wherein each zone comprises a range of logical block addresses;

determining from the logical block address and count information associated with a storage access request, the identity of zones affected by the storage access request; and

,

setting a bit in the persistent record corresponding to the identified zone before completing the storage access request.

11. A storage controller comprising:

mechanisms for handling storage access operations that specify one or more logical block addresses;

mechanisms for accessing metadata indicating groups of logical block addresses that have been previously accessed, and comprising;

a shift register operable to shift the at least one logical block address by a predetermined number of bits based on the size of a logical volume to which the storage access operations apply; and

,

a decoder operable on a predetermined number of the most significant bits of the shifted logical block address to generate a bit map indicating a range of logical block addresses including the at least one logical block address specified by the storage access operation; and

,

mechanisms for skipping selected storage access operations when the selected storage access operations operate on a logical block address that has not been previously accessed.

12. The storage controller of claim 11, wherein the storage access operation comprises a write operation and the mechanisms for accessing metadata further comprise:

processes for updating the metadata to indicate that the write operation has accessed a range of logical block addresses including the logical block address specified in the write operation.

13. The storage controller of claim 11, wherein the storage access operation comprises a read operation and the mechanisms for accessing metadata further comprise:

processes for updating the metadata to indicate that the read operation has accessed a range of logical block addresses including the logical block address specified in the read operation.

14. The storage controller of claim 11, wherein the mechanisms for accessing metadata further comprise:

mechanisms for reading metadata stored persistently in reserved sectors of a physical disk drive.

15. The storage controller of claim 11, wherein the storage controller further comprises non-volatile memory and the mechanisms for accessing metadata further comprise:

mechanisms for reading metadata stored persistently in the non-volatile memory.

16. The storage controller of claim 11 further comprising processes for selectively disabling the mechanisms for skipping.

17. The storage controller of claim 11 further comprising a mechanism for logically combining the bit map with the metadata to produce updated metadata.

18. A method comprising:

tracking storage access operations to a logical disk;

persistently identifying storage blocks that have been accessed; and

,

responsive to the identifying, selectively skipping subsequent operations to storage blocks that have not been accessed.

19. The method of claim 18, wherein the skipping subsequent operations comprises skipping subsequent operations which rely upon previously written storage blocks.

20. The method of claim 18, wherein the skipping subsequent operations comprises one or more of:

skipping a range of LBAs (logical block addresses) when rebuilding the logical disk, skipping a range of LBAs when performing capacity expansion of the logical disk, and skipping a range of LBAs when performing a RAID (redundant array of independent disks) migration operation of the logical disk.