Data block based backup

• US 9,747,168 B2
• Filed: 09/02/2015
• Issued: 08/29/2017
• Est. Priority Date: 10/15/2014
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Haustein, Nils, Mueller-Wicke, Dominic, Prause, Thomas, Steinhauer, Sabine U., Uhl, Andreas
• Primary Examiner(s): Manoskey, Joseph D
• Application Number:
  US14/843,010
• Publication Number:
  US 20160110258A1
• Time in Patent Office: 727 Days
• Field of Search: 714 18, 714 411, 714 63, 714 42
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1469 Backup restoration techniques
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 16/1727 Details of free space management performed by the file system saving storage space on storage systems G06F3/0608; management of blocks in storage devices G06F3/064
• CPC Class Codes: G06F 2201/80 Database-specific techniques
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

The present invention relates to a data block based backup method for a data management system. The data management system comprises a file system that controls access by a database application to at least one database container file stored the data management system. The data management system further comprises a backup client that is connected to a remote backup server, whereby a first version of the database container file is saved in the backup server and a first inode containing information on data blocks of the first version of the database container file. The method may include creating a change tracking table for at least the database container file and adding an entry in the change tracking table, whereby the entry has an indication of the respective data block in association with an information indicating the type of the access.

Claims

1. A data block based backup method for a data management system, the data management system comprising a file system having a database management application program interface (DMAPI), wherein the file system controls access by a database application to at least one database container file set with a DMAPI event mask and stored in a first storage of the data management system, the data management system further comprising a backup client that is connected to a remote backup server in a client-server configuration, wherein a first version of the database container file is saved in the backup server, a first inode comprising information on data blocks of the first version of the database container file, the method comprising:

creating a change tracking table for at least the database container file, and repeatedly adding an entry in the change tracking table in response to detecting an access of the database application through the DMAPI to a data block of the database container file set with the DMAPI event mask, wherein the entry comprises an indication of the respective data block in association with an information indicating the type of the access, wherein the type of the access comprises at least one of a read and a write access tracked by the DMAPI event mask;

in response to receiving a request to backup the database container file;

pausing the execution of the database application and performing a snapshot of a storage volume containing at least the file system and the database container file in a second storage of the data management system;

resuming the database application;

mounting the snapshot storage volume;

recreating the file system including a copy of the database container file in the second storage using the mounted storage volume and recreating the DMAPI event mask associated with the database container file;

identifying entries in the change tracking table referencing the database container file and that are associated with information indicating a write access;

copying the data associated with data blocks of the identified entries to the backup server from the copy of the database container file, thereby backing up a second version of the database container file; and

updating the first inode for indicating the backed up data blocks in the backup server, thereby obtaining a second inode.

2. The method of claim 1, further comprising:

generating a metadata image of the file system including information on the database container file; and

sending the metadata image to the backup server.

3. The method of claim 2, further comprising:

detecting a disaster event in the data management system;

pausing the database application;

retrieving the metadata image of the file system and the second inode from the backup

recreating the file system and an empty database container file using the metadata image; and

resuming the database application, wherein in case of a read access by the database application to a given data block of the database container file, retrieving the given data block from the backup server using the second inode.

4. The method of claim 2, further comprising:

detecting a disaster event in the data management system, wherein the disaster event induces an interruption of the execution of the database application;

retrieving the change tracking table, the metadata image of the file system and the second inode from the backup server;

recreating the file system and an empty database container file using the metadata image;

retrieving data blocks associated with entries having a write and read access type of the database container file in the change tracking table; and

resuming the database application, wherein in case of a read access by the database application to a non-retrieved data block of the database container file, retrieving that data block from the backup server using the second inode.

5. The method of claim 1, further comprising:

receiving a request to restore the second version of the first database file container; and

upon detecting a read access of the data base application to a data block of the current version of the database container file, using the second inode for retrieving the data block from the backup server.

6. The method of claim 1, further comprising:

receiving a request to restore the second version of the first database file container;

retrieving the change tracking table from the backup server;

identifying entries of the change tracking table associated with a write access and entries associated with a read access in case the second version is not the last version;

using the second inode for retrieving the data blocks associated with the identified entries; and

upon detecting a read access of the data base application to a non-retrieved data block of the current version of the database container file, using the second inode for retrieving that data block from the backup server.

7. The method of claim 1, wherein an entry of the tracking change table having information indicating a write or read access is updated only if the corresponding data block is accessed for writing.

8. The method of claim 1, wherein the file system is a general parallel file system (GPFS).

9. The method of claim 1, further comprising:

disabling the entries of the change tracking table that have been backed up by at least one of deleting said entries; and

changing the information to indicate a dummy value for the access type.

10. The method of claim 1, wherein the change tracking table is stored in one of the data management system and the backup server.

11. A computer system for data block based backup for a data management system, the data management system comprising a file system having a database management application program interface (DMAPI), wherein the file system controls access by a database application to at least one database container file set with a DMAPI event mask and stored in a first storage of the data management system, the data management system further comprising a backup client that is connected to a remote backup server in a client-server configuration, wherein a first version of the database container file is saved in the backup server, a first inode comprising information on data blocks of the first version of the database container file, comprising:

one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising;

creating a change tracking table for at least the database container file, and repeatedly adding an entry in the change tracking table in response to detecting an access of the database application through the DMAPI to a data block of the database container file set with the DMAPI event mask, wherein the entry comprises an indication of the respective data block in association with an information indicating the type of the access, wherein the type of the access comprises at least one of a read and a write access tracked by the DMAPI event mask;

in response to receiving a request to backup the database container file;

pausing the execution of the database application and performing a snapshot of a storage volume containing at least the file system and the database container file in a second storage of the data management system;

resuming the database application;

mounting the snapshot storage volume;

recreating the file system including a copy of the database container file in the second storage using the mounted storage volume and recreating the DMAPI event mask associated with the database container file;

identifying entries in the change tracking table referencing the database container file and that are associated with information indicating a write access;

copying the data associated with data blocks of the identified entries to the backup server from the copy of the database container file, thereby backing up a second version of the database container file; and

updating the first inode for indicating the backed up data blocks in the backup server, thereby obtaining a second inode.

12. The computer system of claim 11, further comprising:

generating a metadata image of the file system including information on the database container file; and

sending the metadata image to the backup server.

13. The computer system of claim 12, further comprising:

detecting a disaster event in the data management system;

pausing the database application;

retrieving the metadata image of the file system and the second inode from the backup server;

recreating the file system and an empty database container file using the metadata image; and

resuming the database application, wherein in case of a read access by the database application to a given data block of the database container file, retrieving the given data block from the backup server using the second inode.

14. The computer system of claim 12, further comprising:

detecting a disaster event in the data management system, wherein the disaster event induces an interruption of the execution of the database application;

retrieving the change tracking table, the metadata image of the file system and the second inode from the backup server;

recreating the file system and an empty database container file using the metadata image;

retrieving data blocks associated with entries having a write and read access type of the database container file in the change tracking table; and

resuming the database application, wherein in case of a read access by the database application to a non-retrieved data block of the database container file, retrieving that data block from the backup server using the second inode.

15. The computer system of claim 11, further comprising:

receiving a request to restore the second version of the first database file container; and

upon detecting a read access of the data base application to a data block of the current version of the database container file, using the second inode for retrieving the data block from the backup server.

16. The computer system of claim 11, further comprising:

receiving a request to restore the second version of the first database file container;

retrieving the change tracking table from the backup server;

identifying entries of the change tracking table associated with a write access and entries associated with a read access in case the second version is not the last version;

using the second inode for retrieving the data blocks associated with the identified entries; and

upon detecting a read access of the data base application to a non-retrieved data block of the current version of the database container file, using the second inode for retrieving that data block from the backup server.

17. The computer system of claim 11, wherein an entry of the tracking change table having information indicating a write or read access is updated only if the corresponding data block is accessed for writing.

18. A computer program product for data block based backup for a data management system, the data management system comprising a file system having a database management application program interface (DMAPI), wherein the file system controls access by a database application to at least one database container file set with a DMAPI event mask and stored in a first storage of the data management system, the data management system further comprising a backup client that is connected to a remote backup server in a client-server configuration, wherein a first version of the database container file is saved in the backup server, a first inode comprising information on data blocks of the first version of the database container file, comprising:

one or more computer-readable storage medium and program instructions stored on at least one of the one or more tangible storage medium, the program instructions executable by a processor, the program instructions comprising;

program instructions to create a change tracking table for at least the database container file, and repeatedly adding an entry in the change tracking table in response to detecting an access of the database application through the DMAPI to a data block of the database container file set with the DMAPI event mask, wherein the entry comprises an indication of the respective data block in association with an information indicating the type of the access, wherein the type of the access comprises at least one of a read and a write access tracked by the DMAPI event mask;

program instructions to respond to receiving a request to backup the database container file;

pausing the execution of the database application and performing a snapshot of a storage volume containing at least the file system and the database container file in a second storage of the data management system;

program instructions to resume the database application;

program instructions to mount the snapshot storage volume;

program instructions to recreate the file system including a copy of the database container file in the second storage using the mounted storage volume and recreate the DMAPI event mask associated with the database container file;

program instructions to identify entries in the change tracking table referencing the database container file and that are associated with information indicating a write access;

program instructions to copy the data associated with data blocks of the identified entries to the backup server from the copy of the database container file, thereby backing up a second version of the database container file; and

program instructions to update the first inode for indicating the backed up data blocks in the backup server, thereby obtaining a second inode.

19. The computer program product of claim 18, further comprising:

program instructions to generate a metadata image of the file system including information on the database container file; and

program instructions to send the metadata image to the backup server.

20. The computer program product of claim 19, further comprising:

program instructions to detect a disaster event in the data management system;

program instructions to pause the database application;

program instructions to retrieve the metadata image of the file system and the second inode from the backup server;

program instructions to recreate the file system and an empty database container file using the metadata image; and

program instructions to resume the database application, wherein in case of a read access by the database application to a given data block of the database container file, retrieving the given data block from the backup server using the second inode.