Building a synchronized target database

• US 7,974,943 B2
• Filed: 10/30/2008
• Issued: 07/05/2011
• Est. Priority Date: 10/30/2008
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Smith, Gary S., Fishler, Leonard R., Cohen, Ronald P., Broeder, Sean L., Gilbert, Gary M.
• Primary Examiner(s): Coby; Frantz
• Application Number:
  US12/261,055
• Publication Number:
  US 20100114831A1
• Time in Patent Office: 978 Days
• Field of Search: 707/648, 707609-639
• US Class Current: 707/610
• 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

Abstract

One embodiment is a method that builds a target database with transaction logs from an online source database. The transaction logs include both existing data in the source database and updates that occur to the source database while the target database is being built.

Claims

1. A method for building a database, comprising:

selecting objects in a source database to be copied to a backup database;

beginning to copy the objects to the backup database while the source database remains online;

receiving updates to the source database while the objects are being copied to the backup database; and

applying the updates to both the source database and the backup database while the objects are being copied to the backup database.

2. The method for building a database of claim 1 further comprising:

creating tables in the backup database for the objects;

immediately applying transaction logs that include the updates to the tables to avoid accumulation of the transaction logs while the objects are being copied to the backup database.

3. The method for building a database of claim 2 further comprising:

using the transaction logs to make a fuzzy copy;

generating and applying an audit to bring the fuzzy copy up to date.

4. The method for building a database of claim 1 further comprising:

applying audit to a backup table created in the backup database;

ignoring errors caused by the audit having changes to rows that do not exist in the backup table while the objects are being copied to the backup database to allow the updates to the source database to be immediately entered at the backup database.

5. The method for building a database of claim 1 further comprising:

creating tables in the backup database for the objects;

ignoring errors occurring from deletes and updates to rows that do not exist in the tables.

6. A tangible computer readable storage medium having instructions for causing a computer to execute a method, comprising:

building a target database from an online source database;

receiving updates to the source database while the target database is being built; and

applying records in transaction logs to the target database while the target database is being built, wherein the records in the transaction logs include both existing data in the source database and the updates that occur to the source database while the target database is being built.

7. The tangible computer readable storage medium of claim 6 further comprising, suppressing errors generated when the updates in the transaction logs are applied to tables in the target database and ignoring duplicate record errors for inserts of records that already exist.

8. The tangible computer readable storage medium of claim 6 further comprising, interleaving the updates in the transaction logs as the updates arrive at the source database.

9. The tangible computer readable storage medium of claim 6 further comprising, continuously updating the target database with changes to records that occur to the source database while the target database is being built so the target and source databases are synchronized.

10. The tangible computer readable storage medium of claim 6 further comprising, synchronizing the updates with both the target database and the source database.

11. A database management system, comprising:

an online primary database including a storage device and a controller; and

a replication engine that;

builds a backup database from the online primary database by transferring transaction logs to a backup system that includes the backup database;

receives updates to records in the online primary database;

applies the updates simultaneously to both the online primary database and the backup database while the backup database is being built from the online primary database;

wherein the transaction logs include both data existing in the online primary database and the updates received for the online primary database while the backup database is being built.

12. The database management system of claim 11, wherein the replication engine further:

suppresses errors caused by applying the updates to rows that do not exist in the backup database, wherein the replication engine is informed when synchronization between the online primary database and the backup database starts and finishes for any given object so the replication engine knows when to begin suppressing the errors.

13. The database management system of claim 11 further comprising, a table revive manager (TRM) that ensures tables and indexes are synchronized in both the online primary database and backup database.

14. The database management system of claim 11, wherein the replication engine further:

applies audit to a backup table created in the backup database;

ignores deletes and updates to rows that do not exist in the backup table while data is being copied to the backup database to allow the updates to the online primary database to be immediately entered at the backup database.

15. The database management system of claim 11, wherein the replication engine further:

applies audit to a backup table created in the backup database while ignoring audit having deletes and updates for rows that do not exist in the backup table.

16. The database management system of claim 11, wherein the replication engine applies audit to a backup table created in the backup database, and the audit having revive-inserts corresponding to rows in the backup database that already exist is discarded since the revive-inserts were inserted prior to a point when the audit having the revive-inserts was generated.