Methods and systems for file replication utilizing differences between versions of files

• US 7,844,580 B2
• Filed: 08/14/2007
• Issued: 11/30/2010
• Est. Priority Date: 03/28/2003
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Novell Incorporated (Open Text Corporation)
• Inventors: Murthy, Madhusudhana Honnuduke Srinivasa, Srivastava, Piyush Kumar
• Primary Examiner(s): Jalil; Neveen Abel
• Assistant Examiner(s): RADTKE, MARK A
• Application Number:
  US11/891,962
• Publication Number:
  US 20070288533A1
• Time in Patent Office: 1,204 Days
• Field of Search: 707/690
• US Class Current: 707/690
• CPC Class Codes: G06F 16/1844 Management specifically adapted to replicated file systems
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99953 Recoverability

Abstract

Methods and systems for efficient file replication are provided. In some embodiments, one or more coarse signatures for blocks in a base file are compared with those coarse signatures for blocks of a revised file, until a match is found. A fine signature is then generated for the matching block of the revised file and compared to a fine signature of the base file. Thus, fine signatures are not computed unless a coarse signature match has been found, thereby minimizing unneeded time-consuming fine signature calculations. Methods are also provided for determining whether to initiate a delta file generation algorithm, or whether to utilize a more efficient replication method, based upon system and/or file parameters. In accordance with additional embodiments, the lengths of valid data on physical blocks are obtained from physical block mappings for the files, and these lengths and mappings are utilized for delta file generation, to minimize unnecessary signature computations.

Claims

1. A computer implemented method for comparing data of a first and a revised version of a file to determine differences between the versions, comprising:

segmenting the first and the revised versions into blocks of data of equal size; and

moving a reference frame of a set resolution across adjacent portions of one of the blocks of the revised version to obtain plural coarse signatures of the one of the blocks of the revised version for comparing to a coarse signature of a comparable one of the blocks of the first version;

recursively moving the reference frame by shifting the reference frame in increments of one byte or one bit until a match of the comparing of the coarse signatures is found, if possible; and

upon finding the match, comparing a fine signature of the one of the blocks of the revised version to a fine signature of the comparable one of the blocks of the first version.

2. The method of claim 1, further including creating a delta file upon finding a match between the fine signatures of the revised and first versions.

3. The method of claim 1, wherein the moving the reference frame further includes bridging the reference frame over a boundary of at least one of the blocks of data of the revised version.

4. The method of claim 1, wherein the comparing the fine signatures of the revised and first versions further includes calculating the fine signature of the one of the blocks of the revised version.

5. The method of claim 4, wherein the calculating further includes obtaining a number of bits or bytes adjacent a present position of the reference frame, wherein the number is equal to said equal size of the blocks of data of the revised version.

6. The method of claim 5, wherein after the obtaining, performing a cyclic redundancy check based on the number of bits or bytes adjacent said present position.

7. The method of claim 1, wherein the moving the reference frame further includes counting a shift amount of the reference frame between a first and second position.

8. The method of claim 7, further including using the shift amount in creating a delta file.

9. A computer implemented method for comparing data of a first and a revised version of a file to determine differences between the versions, comprising:

segmenting the first and the revised versions into blocks of data of equal size;

moving a reference frame of a set resolution across adjacent portions of one of the blocks of the revised version to obtain plural coarse signatures of the one of the blocks of the revised version for comparing to a coarse signature of a comparable one of the blocks of the first version, the set resolution of the reference frame being a number of bits less than a second number of bits defining said equal size of the blocks of data;

recursively moving the reference frame until a match of the comparing of the coarse signatures is found, if possible; and

upon finding the match, further including calculating a fine signature of some of the data of the revised version.

10. The method of claim 9, further including comparing the fine signature to a fine signature of the comparable one of the blocks of the first version.

11. The method of claim 9, wherein the calculating the fine signature further includes obtaining a number of bits or bytes adjacent a present position of the reference frame.

12. The method of claim 9, wherein the moving the reference frame further includes bridging the reference frame over a boundary of at least one of the blocks of data of the revised version.

14. The method of claim 9, wherein the moving the reference frame further includes counting a shift amount of the reference frame between various positions.

15. The method of claim 14, further including creating a delta file indicative of the differences between the versions, the creating using the shift amount.

13. The method of 12, wherein the bridging the reference frame further includes accessing data in different memory locations.