Database backup system and method utilizing numerically identified files for incremental dumping

• US 6,374,267 B1
• Filed: 08/13/1999
• Issued: 04/16/2002
• Est. Priority Date: 08/13/1999
• Status: Expired due to Term

• Current Assignee: Unisys Corporation
• Original Assignee: Unisys Corporation
• Inventors: Tam, Marge
• Primary Examiner(s): Breene, John
• Assistant Examiner(s): Le, Debbie M
• Application Number:
  US09/374,352
• Time in Patent Office: 977 Days
• Field of Search: 707/204, 707/205, 707/203
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1456 Hardware arrangements for backup
• CPC Class Codes: G06F 11/1458 Management of the backup or restore process
• CPC Class Codes: G06F 2201/80 Database-specific techniques
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

A method for copying source database files for backup to a destination medium via incremental dump commands based on selecting files from a numerical identification system for spotting updated source files which have not yet been copied and dumped. The optimum block size for dumping onto the destination medium is controlled by the User of the system.

Claims

1. A method for selecting data blocks, in a source database, which have been modified but not yet Dumped, as copies into a destination back-up medium, comprising the steps of:

(a) generating a data block identifier word, which is independent of time-stamps, for placement in each data block that was modified after a selected target date and time;

(b) utilizing a utility program to select copying of said modified data blocks having an identifier word indicating that it was modified after said selected target date and time;

(c) placing said selected copies of said modified data blocks onto said destination medium.

2. The method of claim 1 wherein step (a) includes the steps of:

(a1) utilizing an ACCESSROUTINES program having a numerical generator to generate a time-identifier word, designated as a transtamp, for placement into each of said data blocks in said Source database in order to indicate the time sequence in which each of said data blocks was modified.

3. The method of claim 1 wherein step (b) includes the step of:

(b1) invoking an Incremental Dump command to initiate copying of modified data blocks having a transtamp greater than a selected target date and time.

4. The method of claim 1 wherein step (c) includes the step of:

(c1) adjusting the block size of the selected modified blocks to provide a more compatible block size suitable to said destination medium.

5. A method for copying data blocks from a Source database, onto a destination medium which is independent of the normally-used time stamps on each data block in said Source data base, comprising the steps of:

(a) utilizing a control file means to provide a DUMPSTAMP operation which enables a unique sequential identification number for each data block in said Source database;

(b) invoking an access routine to initiate a numerical generator for placing said unique sequential identification number, designated as a transtamp, in each data block which is modified after a certain target date;

(c) initiating an Accumulated Dump command for reading those modified data blocks having a DUMPSTAMP value greater than the last full DUMPSTAMP value in said control file through use of said unique transtamp identification number which selects those modified data blocks which were modified after said target date and time;

(d) selecting the modified data blocks according to their transtamp identification numbers, for copying and placement on said destination medium.

6. A system for dumping copies of modified data blocks from a source database by selecting data blocks according to a numerical identifier for each modified data block, which numerical identifier is independent of date and time stamps, comprising:

(a) means for identifying modified data blocks, in a source data base, said means being independent of a date-time stamp;

(b) means for selecting a Dump of said modified data blocks for copying onto a destination medium according to those data blocks which were only modified after the completion of the last full or incremental Dump operation.

7. The system of claim 6 which includes:

(c) means for selecting a Dump of modified data blocks which were modified after a selected target date and time.

8. The system of claim 7 wherein said means for identifying modified data blocks includes:

(a1) ACCESSROUTINES means for initiating a numerical generator means to generate a time-identifier word for placement into each of said data blocks in said database, in order to indicate the time sequence in which each of said data blocks was modified.

9. The system of claim 7 wherein said means for selecting includes:

(c1) means for invoking a utility program with an Incremental Dump command to initiate copying of said data blocks from said Source database to a destination medium by selecting only those modified data blocks which have a sequence identifier word which is greater than a selected target data and time.

10. The system of claim 7 which includes:

(d) means for copying said selected blocks of data files from said Source database for placement onto said destination medium.

11. The system of claim 10 wherein said means for copying includes:

(d1) means for adjusting the block size of said selected blocks being dumped onto said destination medium in order to provide a compatible block size suitable to said destination medium.

12. A system for selecting modified data blocks in a source database for copying and dumping to a destination medium which selection of modified blocks is independent of any form of time stamps attributable to each data block in said Source database, comprising:

(a) control file means providing a DUMPSTAMP for each structured data set consisting of data blocks in said Source data base and storing a transtamp of the last full dump;

(b) ACCESSROUTINE means for initiating a numerical generator to place a unique identifying number, designated as a transtamp, in each data block in said source database;

(c) means to initiate an incremental Dump command to Dump copies of modified data blocks onto a destination medium;

(d) DUMPSTAMP means for calculating said transtamp value for each modified data block to identify it as being modified after the date of the last Dump operation;

(e) placing said modified data blocks onto said destination medium.