On-line database duplication with incorporation of concurrent database updates

• US 6,023,707 A
• Filed: 12/30/1997
• Issued: 02/08/2000
• Est. Priority Date: 01/16/1997
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Ishihara, Kazuo, Fujita, Tsuyoshi, Kawashima, Masayuki, Takeda, Yoshihisa, Masumoto, Toshihiko, Masatomi, Katsutoshi, Hamada, Yukihiro
• Primary Examiner(s): HOMERE, JEAN RAYMOND
• Application Number:
  US09/000,809
• Time in Patent Office: 770 Days
• Field of Search: 707/201, 707/202, 707/203, 707/204, 395/675, 395/676, 395/182.09, 395/182.04, 395/182.11
• US Class Current: 707/648
• CPC Class Codes: G06F 11/1402 Saving, restoring, recovering or retrying
• 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: G06F 16/275 Synchronous replication
• CPC Class Codes: Y10S 707/99953 Recoverability

Abstract

A system is provided for duplication of an update database in a computer using partial data extraction. A copying module creates a copy database, including a partial copy, of the update database. A journal selecting module selects data, related to parts of the copy database in which the copy has finished, from an update data journal of the update database that is made when the copy database is being copied by the copying module. A duplication module updates the contents of the copy database created by the copying module with the contents of the update data journal selected by the journal selecting module.

Claims

1. A system for duplicating an update database in a computer during online processing, comprising:

a first unit to create at least a copy database by logically copying contents in the update database, in sequence on a block-by-block basis in accordance with a specification of a desired extraction range, during online processing, regardless of whether the physical data structure of the update database is not the same as that of the copy database;

a second unit to create selected journal data by recording an update data journal of the update database caused by transactions while copying the contents of the update database in a journal file, and by selecting data relating to parts of the copy database in which the copy has finished from the update data journal; and

a third unit to create a duplicate database, from the copy database created by the first unit and the selected journal data created by the second unit.

2. A system for duplicating an update database according to claim 1, wherein database processing in the first, second, and third units is performed based on logical data structure information of the database.

3. A system for duplicating an update database according to claim 2, wherein a data range for the copying in the first unit is specified externally by the logical data structure information.

4. A system for duplicating an update database according to claim 2, wherein in the second unit, the selection of the update data is performed by using position information.

5. A system for duplicating an update database according to claim 2, wherein in the second unit, the selection of the update data is performed by using time information.

6. A system for duplicating an update database according to claim 2, further comprising a record position correspondence table for managing correspondences between record position information in the update database and record position information in the duplicate database.

7. A system for duplicating an update database according to claim 6, wherein the record position correspondence table is held by selecting either a main memory or a nonvolatile medium for storage thereof.

8. A system for duplicating an update database according to claim 1, further comprising:

a collective input unit for collectively inputting a plurality of journals from the update data journal;

a sequential input unit for sequentially inputting journals one at a time from the update data journal;

a journal input switching unit for dynamically switching between the collective input unit and the concurrent input unit; and

a mode switching unit for outputting a control journal for directing input mode switching.

9. A system for duplicating an update database according to claim 8, wherein the input mode switching is effected according to an initial mode specified externally by a control statement, an operating time segment specified externally by a control statement, or an online processing load.

10. A recording medium recording a computer program for duplicating an update database in a computer during online processing the computer program comprising the steps of:

creating at least a copy database by logically copying contents in the update database, in sequence on a block-by-block basis in accordance with a specification of a desired extraction range, during online processing, regardless of whether the physical data structure of the update database is not the same as that of the copy database;

selecting journal data by recording an update data journal of the update database caused by transactions while copying the contents of the update database in a journal file, and by selecting data relating to parts of the copy database in which the copy has finished from the update data journal; and

creating a duplicate database, from the copy database created in the copy database creating step and the selected journal data created in the journal data selecting step.

11. An update database duplicating method, comprising:

creating a copy database during online processing by logically copying contents in an update database, in sequence on a block-by-block basis in accordance with a specification of a desired extraction range, a physical data structure of the copy database being independent of a physical data structure of the update database;

creating selected journal data by recording an update data journal for transactions while copying the contents of the update database, and by selecting from the update data journal data relating to parts of the copy database for which the copying has finished; and

creating a duplicate database by modifying the copy database in accordance with the selected journal data.

12. An update database duplicating method according to claim 11, wherein the copy database creating process, the selected journal data creating process, and the duplicate database creating process are performed based on logical data structure information of the update database.

13. An update database duplicating method according the claim 12, wherein database processing in the copy database creating operation, the selected journal data creating operation, and the duplicate database creating operation is performed based on logical data structure information of the update database.

14. An update database duplicating method according to claim 13, wherein a data range for the copying operation is specified externally by the logical data structure information.

15. An update database duplicating method according to claim 13, wherein the selected journal data is selected using position information.

16. An update database duplicating method according to claim 13, wherein the selected journal data is selected using time information.

17. An update database duplicating method according to claim 13, further comprising managing correspondences between record position information in the update database and record position information in the duplicate database using a record position correspondence table.

18. An update database duplicating method according to claim 17, wherein the record position correspondence table is maintained by selecting a main memory or a nonvolatile medium for storage thereof.

19. An update database duplicating method according to claim 12, further comprising:

collectively inputting a plurality of journals from the update data journal responsive to receipt of a collective input instruction;

sequentially inputting journals one at a time from the update data journal responsive to receipt of a sequential input instruction;

dynamically switching between collective input and sequential input responsive to input mode switching control; and

outputting a control journal for controlling the input mode switching by a collective input instruction or a sequential input instruction.

20. An update database duplicating method according to claim 19, wherein the input mode switching is carried out according to an initial mode specified externally by a control statement, an operating time segment specified externally by a control segment, or an online processing load.