Dynamic rewrite of files within deduplication system

• US 8,433,690 B2
• Filed: 06/04/2012
• Issued: 04/30/2013
• Est. Priority Date: 12/01/2010
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Aronovich, Lior, Levy, Asaf, Schreiber, Amit, Krikler, Samuel
• Primary Examiner(s): Jalil, Neveen Abel
• Assistant Examiner(s): BUCKINGHAM, KELLYE DEE
• Application Number:
  US13/488,164
• Publication Number:
  US 20120239633A1
• Time in Patent Office: 330 Days
• Field of Search: 707/692
• US Class Current: 707/692
• 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

Abstract

A dynamic layer above a sequential deduplication file system (denoted as DFS) implements the rewrite functionality. A user file is composed of one or more DFS files. As incoming data is written into a user file, the data is written by the dynamic layer sequentially into DFS files, created one by one. For each user file this dynamic layer creates and maintains a dynamic metadata file, in a regular, non deduplicated file system. This metadata file contains entries pointing to sections of DFS files.

Claims

1. A method for rewriting data in a deduplication storage environment by a processor device, comprising:

creating a plurality of new deduplication file system (DFS) files to be stored as part of a plurality of user files, the plurality of user files including original deduplication file system (DFS) files and being accessible by a sequential deduplication file system and a dynamic non-deduplicated file system, the dynamic non-deduplicated file system including a plurality of dynamic metadata files having entries pointing to corresponding sections of the original DFS files each of the plurality of new DFS files including rewritten data resulting from rewrite operations;

grouping the rewrite operations into a set of files having rewritten sections and a set of files having unaffected data;

directly writing rewritten sections of one DFS file of the set of files having rewritten sections into an other new DFS file;

quoting unmodified sections of the one of the DFS files of the set of files having rewritten section into the other new DFS file, and

deleting the original DFS files corresponding to the one of the DFS files of the set of files having rewritten sections.

2. The method of claim 1, wherein quoting the unmodified sections of the one of the DFS files of the set of files having rewritten section may be performed into additional new DFS files.

3. The method of claim 1, wherein the deduplication file system is static.

4. The method of claim 1, wherein the entries of one of the dynamic metadata files may be modified to reflect a new layout of one of the user files over the DFS files.

5. The method of claim 4, further comprising replacing entries in a dynamic metadata file associated with the deleted one of the DFS files of the set of files having rewritten section with new entries pointing to the new DFS.

6. The method of claim 1, further comprising using a dynamic file system layer to implement the sequential deduplication file system and the dynamic non-deduplicated file system to create and maintain the user files.

7. The method of claim 1, wherein the step of using a dynamic file system layer includes sequentially composing the DFS files of the user files.

8. The method of claim 7, wherein the data within one of the sequentially composed DFS files is non-sequential in terms of the user file.