Granted Patent B2

US 7,584,341 · App. 11/544,213 · Granted Sep 1, 2009

Abstract

Claims (60)

1. A method for storage optimization by de-fragmenting virtual volumes of a storage area network (SAN), comprising the steps of:

collecting statistics regarding a performance of each of a plurality of virtual volumes;

determining which of the plurality of virtual volumes in the SAN to optimize;

assessing a de-fragment policy for the virtual volume to be optimized by:

determining, using a mapping table, data fragments in each logical unit to be migrated;

determining, using the mapping table, data fragments in each LU in the virtual volume to be combined; and

determining, using the mapping table data fragments in each LU in the virtual volume that cannot be altered;

determining for each virtual volume to be optimized, a new target location for one or more logical units in the virtual volume; and

transferring data blocks of the one or more logical units to the new target location.

2. The method of claim 1 , wherein the step of collecting statistics comprises a step of: determining at least one of: a number of input/output operations per second for the virtual volume; fragments in the virtual volume; a number of routers and switches in the path between each host and each physical storage device; and a class of storage of the virtual volume.

3. The method of claim 2 , wherein the determination step is based on at least the class of storage of the virtual volumes.

4. The method of claim 3 , wherein the determination step is performed by comparing the collected statistics to a predetermined baseline value.

5. The method of claim 1 , wherein the virtual volumes to be optimized are virtual volumes that can be de-fragmented.

6. The method of claim 1 , wherein the data fragments are comprised of data blocks belonging to logical units of the respective virtual volumes.

7. The method of claim 1 , wherein the mapping table includes mapping information on virtual volumes and logical units to their respective physical storage devices.

8. The method of claim 1 , wherein determining the new target location for each logical unit in the virtual volume comprises:

computing a physical address of the new location target; and updating the mapping table with the computed physical address.

9. A computer readable storage medium for storing a code executed by a computer, said code enabling data storage optimization by de-fragmenting virtual volumes of a storage area network, and performing the steps of:

collecting statistics regarding to a performance of each of the virtual volumes;

determining which of the virtual volumes in the SAN to optimize;

assessing a de-fragment policy for the virtual volume to be optimized by:

determining, using a mapping table, data fragments in each logical unit to be migrated;

determining, using the mapping table, data fragments in each LU in the virtual volume to be combined; and

determining, using the mapping table data fragments in each LU in the virtual volume that cannot be altered;

determining for each virtual volume to be optimized a new target location for each logical unit in the virtual volume; and

transferring data blocks of the logical unit to the new target location.

10. The computer readable storage medium of claim 9 , wherein collecting the statistics comprises a step of: determining at least one of: a number of input/output (I/O) operations per second for the virtual volume; fragments in the virtual volume; a number of routers and switches in a path between each host and each physical storage device; and a class of storage of the virtual volume.

11. The computer readable storage medium of claim 10 , wherein the determination step is based on at least the class of storage of the virtual volumes.

12. The computer readable storage medium of claim 11 , wherein the determination step is performed by comparing the collected statistics to a predetermined baseline value.

13. The computer readable storage medium of claim 9 , wherein the virtual volumes to be optimized are virtual volumes that can be de-fragmented.

14. The computer readable storage medium of claim 9 , wherein the data fragments are comprised of data blocks belonging to logical units of respective virtual volumes.

15. The computer readable storage medium of claim 9 , wherein the mapping table includes mapping information on virtual volumes and logical units to their respective physical storage devices.

16. The computer readable storage medium of claim 9 , wherein determining the new target location for each logical unit in the virtual volume comprises:

computing a physical address of the new location target; and

updating the mapping table with the computed physical address.

17. A virtualization switch in combination with a storage area network and capable of optimization virtual volumes in the storage area network, the virtualization switch comprising:

software instructions

a memory for maintaining at least a mapping table;

a storage communication means for communicating with a plurality of physical storage devices;

a network communication means for communicating with a plurality of hosts over a network; and

a processor for executing said software instructions adapted to enable the virtualization switch to perform the operations of:

collecting statistics regarding to a performance of each of the virtual volumes;

determining which of the virtual volumes in the storage area network to be optimized;

assessing a de-fragment policy for the virtual volume to be optimized by:

determining, using a mapping table, data fragments in each logical unit to be migrated;

determining, using the mapping table, data fragments in each LU in the virtual volume to be combined; and

determining, using the mapping table data fragments in each LU in the virtual volume that cannot be altered;

determining for each virtual volume to be optimized a new target location for each logical unit logical unit in the virtual volume; and

transferring data blocks of the logical unit to the new target location.

18. The virtualization switch of claim 17 , wherein collecting the statistics comprises a step of determining at least one of: a number of input/output (I/O) operations per second for the virtual volume; fragments in the virtual volume; a number of routers and switches in a path between each host and each physical storage device; and a class of storage of the virtual volume.

19. The virtualization switch of claim 18 , wherein the determination on the virtual volumes to be optimized is based on at least the class of storage of the virtual volumes.

20. The virtualization switch of claim 19 , wherein the determination step is performed by comparing the collected statistics to a predetermined baseline value.

21. The virtualization switch of claim 17 , wherein the virtual volumes to be optimized are virtual volumes that can be de-fragmented.

22. The virtualization switch of claim 17 , wherein the data fragments comprise data blocks belonging to logical units of respective virtual volumes.

23. The virtualization switch of claim 17 , wherein the mapping table includes mapping information on virtual volumes and logical units to their respective physical storage devices.

24. The virtualization switch of claim 17 , wherein determining the new target location for each logical unit in the virtual volume comprises:

computing a physical address of the new location target; and

updating the mapping table with the computed physical address.

25. The virtualization switch of claim 17 , wherein the physical storage device comprises at least one of: a tape drive, a tape library, an optical drive, a disk, redundant array of independent disks (RAID).

26. The virtualization switch of claim 17 , wherein the network comprises at least one of: a local area network (LAN), a wide area network (WAN).

Assignments (8)