Method and system for backup and recovery

Patent metadata

Field	Value
Publication number	US-9588986-B2
Application number	US-201615052371-A
Country	US
Kind code	B2
Filing date	Feb 24, 2016
Priority date	Jan 2, 2012
Publication date	Mar 7, 2017
Grant date	Mar 7, 2017

Abstract

For data backup and recovery based on linked file repositories with each of the linked file repositories representing an individual file system capable of storing at least one version of a file and being connected to at least one server system, each of the linked file repositories are placed in a certain position for storing a certain version of the file. Each position of each of the linked file repositories is continuously numbered. A number of the versions of the file are determined by the position of the one of the linked file repositories.

First claim

What is claimed is:

1. A method for data backup and recovery by a processor based on linked file repositories with each of the linked file repositories representing an individual file system capable of storing at least one version of a file and being connected to at least one server system, wherein the method includes:
placing each of the linked file repositories in one of a plurality of positions for storing the at least one version of the file;
continuously numbering each of the plurality of positions of each of the linked file repositories;
determining a number of the at least one version of the file by the one of a plurality of positions of the one of the linked file repositories;
for each of the linked file repositories, performing at least one function operation for each file stored in the linked file repositories matching at least one rule, wherein the at least one function operation includes at least one of a deduplication function allowing deduplicating files based on the version number, an indexing function allowing to index the file and providing search and discovery capabilities, and an expiration function for moved files determining when a moved file expires and is being deleted time-based or event-based; wherein the at least one rule includes a file extension matching a certain pattern, a file name matching one of a plurality of patterns, a file path name matching the one of a plurality of patterns, a file owner matching the one of a plurality of patterns, a file creation, and a modification or last access time matching a specified date and a specified time range; and
implementing the version-recovery process on a block-differential basis by performing one of:
receiving a write operation with a new one of the at least one version of the file to be written from one of the at least one server system and a preceding one of the linked file repositories,
determining if the new one of the at least one version of the file to be written exists in the one of the linked file repositories,
if the new one of the at least one version of the file to be written exists in the one of the linked file repositories performing one of:
comparing data of the new one of the at least one version of the file to be written and an existing one of the at least one version of the of the file to be written,
determining different data blocks,

moving the different data blocks from the existing one of the at least one version of the of the file to be written from the one of the linked file repositories to a subsequent one of the linked file repositories if versioning policies indicate a higher number of the at least one version as compared to a position number of the one of the linked file repositories, otherwise:
deleting the different data blocks of the existing one of the at least one version in the one of the linked file repositories, and

storing the different data blocks of the new one of the at least one version of the file to be written from one of the at least one server system and the preceding one of the linked file repositories, and
if the at least one version of the file to be written does not exist in the one of the linked file repositories:
storing the new one of the at least one version of the file to be written from the one of the linked file repositories;
including the version number and a file name of the requested version of the file to be read as part of the read operation, wherein the requested version number is one of an integer number, a date range, and a time range; and
setting the version number of the requested version as a file attribute prior to the read operation.

2. The method of claim 1 , further including implementing a version-recovery process over each of the linked file repositories to perform a read operation of the at least one version of the file.

3. The method of claim 1 , further including continuously numbering each of the plurality of positions of each of the linked file repositories beginning at a numerical value of 1, wherein a first one of the linked file repositories is numbered as 1, and a first version of the at least one version of the file representing a most recent version of the file is stored.

4. The method of claim 1 , further including implementing the version-recovery process by performing one of:
receiving a write operation and a new one of the at least one version of the file to be written from one of the at least one server system and a preceding one of the linked file repositories,
determining if the new one of the at least one version of the file to be written exists in the one of the linked file repositories,
if the new one of the at least one version of the file to be written exists in the one of the linked file repositories performing one of:
verifying versioning policies,

moving an existing one of the at least one version from the corresponding one of the linked file repositories to a subsequent one of the linked file repositories if the versioning policies indicate a higher number of the at least one version that are to be saved as compared to a position number of the one of the linked file repositories, otherwise:
deleting the existing one of the at least one version in the one of the linked file repositories, and

storing the new one of the at least one version of the file to be written from one of the at least one server system and the preceding one of the linked file repositories, and
if the new one of the at least one version of the file to be written does not exist in the one of the linked file repositories:
storing the new one of the at least one version of the file to be written from the one of the linked file repositories.

5. The method of claim 4 , further including implementing the version-recovery process by performing one of:
receiving the read operation and a version number of a requested version of the file to be read from the at least one server system,
examining the version number of the file to be read, and
determining if the requested version of the file to be read exists in one of the linked file repositories having a position number matching the received version number, wherein if the requested version of the file to be read exists:
reading and sending the file to a requesting server system, otherwise

indicating a file read-error to the requesting server system if the requested version of the file to be read does not exist.

6. The method of claim 1 , further including implementing the version-recovery process on the block-differential basis by performing one of:
receiving a read operation and a version number of a requested version of the file to be read from the at least one server system,
examining the version number of the file to be read;
determining if the requested version of the file to be read exists in one of the linked file repositories having the position number matching the received version number,
if the requested version of the file to be read exists, recreating the requested version of the file to be read by determining if the requested version of the file to be read is a most recent version of the at least one version of the file to be read,
if the requested version of the file to be read is the most recent version of the at least one version of the file to be read:
reading and sending the most recent version of the at least one version of the file to be read from a first one of the linked file repositories to a requesting server system, otherwise performing one of:
reading content of the linked file repositories having the position number that is one of less than and equal to the requested version number of the at least one version of the file to be read,
replacing the different data blocks of the most recent version of the at least one version of the file to be read

Assignees

• IBM

Inventors

• Haustein Nils
• Krause Thorsten
• Seipp Harald
• Winarski Daniel J

Classifications

• G06F2201/80

  Database-specific techniques · CPC title

• G06F11/1448 Primary

  Management of the data involved in backup or backup restore · CPC title

• G06F16/162

  Delete operations (erasing in storage systems G06F3/0652) · CPC title

• G06F16/1873

  Versioning file systems, temporal file systems, e.g. file system supporting different historic versions of files · CPC title

• G06F2201/84

  Using snapshots, i.e. a logical point-in-time copy of the data · CPC title