VIRTUAL BLOCK DEVICE

• US 20110276543A1
• Filed: 11/05/2010
• Published: 11/10/2011
• Est. Priority Date: 03/30/2009
• Status: Active Grant

• Current Assignee: Exar Corporation (Maxlinear Incorporated)
• Original Assignee: Exar Corporation (Maxlinear Incorporated)
• Inventors: Matze, John Edward Gerard
• Granted Patent: US 8,127,109 B2
• Time in Patent Office: Days
• US Class Current: 707/692
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 16/1748 De-duplication implemented within the file system, e.g. based on file segments de-duplication techniques in storage systems for the management of data blocks G06F3/0641
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0641 De-duplication techniques
• CPC Class Codes: G06F 3/0674 Disk device

Abstract

A virtual block device is an interface with applications that appears to the applications as a memory device, such as a standard block device. The virtual block device interacts with additional elements to do data deduplication to files at the block level such that one or more files accessed using the virtual block device have at least one block which is shared by the one or more files.

Claims

1. A virtual block device that is an interface with applications, the virtual block device appearing to the applications as a memory device, wherein the virtual block device interacts with additional elements to do data deduplication to files at the block level such that one or more files accessed using the virtual block device have at least one block which is shared by the one or more files, and wherein the virtual block device is implemented using at least one processor.

2. The virtual block device of claim 1, wherein the at least one block is shared by at least two different files.

3. The virtual block device of claim 1, wherein the virtual block device appears to applications as a standard block device.

4. The virtual block device of claim 1, wherein the virtual block device is operable to set a block size for deduplication.

5. A system including the virtual block device of claim 1, further comprising additional virtual block devices and a driver that manages the virtual block device and the additional virtual block devices.

6. A system including the virtual block device of claim 1, wherein the additional elements do the functions of generating a hash value for a block of data, determining if the hash value is unique, discarding the hash value and the block of data if the hash value is not unique;

and writing the block of data to a disk if the hash value is unique.

7. The system of claim 6, further comprising updating a metadata counter for the hash value.

8. A system including the virtual block device of claim 1, wherein the additional elements include a unit to perform hashing operations and compression.

9. The system of claim 8, wherein the unit is a card to perform hardware-assisted compression.

10. The system of claim 8, wherein the additional elements include a deduplication block manager that passes blocks to the unit and receives a hash for each block from the unit.

11. The system of claim 10, wherein the deduplication block manager receives at least some compressed blocks from the unit.

12. The system of claim 11, wherein the additional elements include a truncated hash array that includes an index for each possible truncated hash value and a corresponding entry initially set to a default value wherein when a new hash is processed, the corresponding entry is modified to include a pointer to the new hash and block.

13. The system of claim 12,

wherein the unit

receives blocks from the deduplication block manager,

creates a hash value for each block of data using a hashing algorithm, and

compresses at least some blocks of data using a compression algorithm;

and

wherein the deduplication block manager

receives the hash values and the at least some compressed blocks of data from the unit, and

creates a truncated hash corresponding to each hash value.

14. The system of claim 13, further comprising:

wherein the deduplication block manager looks up each truncated hash in the truncated hash array; and

wherein if the corresponding entry to the truncated hash is the default value then the virtual block device writes the block to a computer readable medium and adds a pointer to the block and hash value to the truncated hash array.

15. The system of claim 13, further comprising:

wherein the deduplication block manager looks up each truncated hash in the truncated hash array; and

wherein if the corresponding entry to the truncated hash includes a pointer to a processed hash value and block then the deduplication block manager retrieves the processed hash value and checks whether the hash value corresponding to the truncated hash matches the processed hash value.

16. The system of claim 15, wherein if the hash value matches the processed hash value then the block is discarded.

17. The system of claim 15, wherein if the hash value does not match the processed hash value then the deduplication block manager determines if there are pointers to additional processed hash values and the additional processed hash values are compared with the hash value.

18. The system of claim 17, wherein if the hash value does not match any processed hash values then the virtual block device writes the block to a computer readable medium and adds a pointer to the block and hash value to the truncated hash array.

19. A computer readable medium including a virtual block device that is an interface with applications, the virtual block device appearing to the applications as a memory device, wherein the virtual block device interacts with additional elements to do data deduplication to files at the block level such that one or more files accessed using the virtual block device have at least one block which is shared by the one or more files.

20. The computer readable medium of claim 19, wherein the at least one block is shared by at least two different files.

21. The computer readable medium of claim 19, wherein the virtual block device appears to applications as a standard block device.

22. The computer readable medium of claim 19, wherein the virtual block device is operable to set a block size for deduplication.

23. The computer readable medium of claim 19, further comprising additional virtual block devices and a driver that manages the virtual block device and the additional virtual block devices.