APPROACH FOR OPTIMIZING RESTORES OF DEDUPLICATED DATA

• US 20110218969A1
• Filed: 03/08/2010
• Published: 09/08/2011
• Est. Priority Date: 03/08/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yakushev, Mark L., Anglin, Matthew J., Hochberg, Avishai H., Cannon, David M.
• Granted Patent: US 8,370,297 B2
• Time in Patent Office: Days
• US Class Current: 707/679
• CPC Class Codes: G06F 11/1402 Saving, restoring, recovering or retrying
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 11/1469 Backup restoration techniques
• CPC Class Codes: G06F 11/2089 Redundant storage control functionality

Abstract

Various techniques for improving the performance of restoring deduplicated data files from a server to a client within a storage management system are disclosed. In one embodiment, a chunk index is maintained on the client that tracks the chunks remaining on the client for each data file that is stored to and restored from the storage server. When a specific file is selected for restore from the storage server to the client, the client determines if any local copies of this specific file'"'"'s chunks are stored in files already existing on the client data store. The file is then reconstructed from a combination of these local copies of the file chunks and chunks retrieved from the storage server. Therefore, only chunks that are not stored or are inaccessible to the client are retrieved from the server, reducing server-side processing requirements and the bandwidth required for data restore operations.

Claims

1. A method for optimizing restores of deduplicated data within a storage management system, the storage management system comprising a client computing system coupled to a client data store and a server computing system coupled to a deduplicated server data store, with the method comprising:

maintaining a chunk index on the client computing system, the chunk index tracking locations of chunks within files remaining on the client data store after storage of the files to the server data store;

retrieving chunks for a specified file during a restore of the specified file from the server data store to the client data store, including for each chunk of the specified file;

determining whether a valid entry for the chunk exists in the chunk index;

retrieving the chunk from the server data store responsive to determining a valid entry for the chunk does not exist in the chunk index; and

retrieving the chunk from a location on the client data store specified in the chunk index responsive to determining a valid entry for the chunk exists in the chunk index; and

reconstructing the specified file from the retrieved copies of the chunks and completing the restore of the specified file to the client data store.

2. The method of claim 1, further comprising updating the chunk index on the client computing system for files restored to the client data store, thereby tracking locations of chunks within reconstructed files on the client data store in the chunk index.

3. The method of claim 1, further comprising updating the chunk index on the client computing system subsequent to retrieval of each chunk from the server data store, thereby enabling subsequent identical chunks within the specified file to be retrieved from the specified file on the client data store.

4. The method of claim 1, wherein the location on the client data store specified in the chunk index includes a file path on the client data store and an offset within the file located at the file path.

5. The method of claim 1, further comprising storing, in the chunk index, file attributes of each file having chunks for all chunks tracked in the chunk index.

6. The method of claim 1, wherein the chunk index contains values used for locating and verifying the chunk in the client data store, including a hash value of the chunk, a length of the chunk, a location of a file containing the chunk, file attributes of the file containing the chunk, and an offset of the chunk within the file containing the chunk.

7. The method of claim 1, wherein determining whether a valid entry for the chunk exists in the chunk index includes validating existence of the chunk in the file at the location on the client data store specified in the chunk index.

8. The method of claim 1, further comprising retrieving the chunk from the server data store responsive to being unable to retrieve the chunk from the location on the client data store specified in the chunk index.

9. The method of claim 1, further comprising checking file attributes provided by the entry in the chunk index against a file stored at the location on the client computing system indicated by the entry to verify that a valid version of the file exists on the client data store.

10. The method of claim 1, further comprising comparing a digest of the reconstructed specified file to a digest stored by the server computing system to verify successful reconstruction of the specified file.

11. A method for optimizing restores of deduplicated data onto a client computing system in a storage management system, the storage management system comprising a client computing system coupled to a client data store and a server computing system coupled to a deduplicated server data store, with the method comprising:

receiving data of a file from the client data store for storage to the server data store;

storing deduplicated chunks of the file within the server data store, wherein the client computing system maintains an index to track locations of file chunks remaining on the client data store after storage to the server data store;

providing one or more chunks of the file from the server data store during a restore of the file to the client data store, including providing the chunk from the server data store responsive to the client computing system determining the one or more chunks do not exist on the client data store; and

providing an end-to-end digest of the file to the client computing system to enable verification of the restore of the file to the client data store.

12. A storage management system, comprising:

a client computing system;

a server computing system connected to the client computing system via a network;

a client data store coupled to the client computing system;

a server data store coupled to the server computing system;

at least one processor within the storage management system; and

at least one memory store within the storage management system having instructions operable with the at least one processor for optimizing restores of deduplicated data, the instructions being executed for;

maintaining a chunk index on the client computing system, the chunk index tracking locations of chunks within files remaining on the client data store after storage of the files to the server data store;

retrieving chunks for a specified file during a restore of the specified file from the server data store to the client data store, including for each chunk of the specified file;

determining whether a valid entry for the chunk exists in the chunk index;

retrieving the chunk from the server data store responsive to determining a valid entry for the chunk does not exist in the chunk index; and

retrieving the chunk from a location on the client data store specified in the chunk index responsive to determining a valid entry for the chunk exists in the chunk index; and

reconstructing the specified file from the retrieved copies of the chunks and completing the restore of the specified file to the client data store.

13. The storage management system of claim 12, further comprising instructions being executed for updating the chunk index on the client computing system for files restored to the client data store, thereby tracking locations of chunks within reconstructed files on the client data store in the chunk index.

14. The storage management system of claim 12, further comprising instructions being executed for updating the chunk index on the client computing system subsequent to retrieval of each chunk from the server data store, thereby enabling subsequent identical chunks within the specified file to be retrieved from the specified file on the client data store.

15. The storage management system of claim 12, wherein the location on the client data store specified in the chunk index includes a file path on the client data store and an offset within the file located at the file path.

16. The storage management system of claim 12, further comprising instructions being executed for storing, in the chunk index, file attributes of each file having chunks for all chunks tracked in the chunk index.

17. The storage management system of claim 12, wherein the chunk index contains values used for locating and verifying the chunk in the client data store, including a hash value of the chunk, a length of the chunk, a location of a file containing the chunk, file attributes of the file containing the chunk, and an offset of the chunk within the file containing the chunk.

18. The storage management system of claim 12, wherein determining whether a valid entry for the chunk exists in the chunk index includes validating existence of the chunk in the file at the location on the client data store specified in the chunk index.

19. The storage management system of claim 12, further comprising instructions being executed for retrieving the chunk from the server data store responsive to being unable to retrieve the chunk from the location on the client data store specified in the chunk index.

20. The storage management system of claim 12, further comprising instructions being executed for checking file attributes provided by the entry in the chunk index against a file stored at the location on the client computing system indicated by the entry to verify that a valid version of the file exists on the client data store.

21. The storage management system of claim 12, further comprising instructions being executed for comparing a digest of the reconstructed specified file to a digest stored by the server computing system to verify successful reconstruction of the specified file.

22. A storage management system, comprising:

a client computing system;

a server computing system connected to the client computing system via a network;

a client data store coupled to the client computing system;

a server data store coupled to the server computing system;

at least one processor within the storage management system; and

at least one memory store within the storage management system having instructions operable with the at least one processor for optimizing restores of deduplicated data from the server computing system onto the client computing system, the instructions being executed for;

receiving data of a file from the client data store for storage to the server data store;

storing deduplicated chunks of the file within the server data store, wherein the client computing system tracks locations of file chunks remaining on the client data store after storage to the server data store;

providing one or more chunks of the file from the server data store during a restore of the file to the client data store, including providing the chunk from the server data store responsive to the client computing system determining the one or more chunks do not exist on the client data store; and

providing an end-to-end digest of the file to the client computing system to enable verification of the restore of the file to the client data store.