Method for sequential coordination of external database application events with asynchronous internal database events

• US 7,426,559 B2
• Filed: 05/09/2002
• Issued: 09/16/2008
• Est. Priority Date: 05/09/2002
• Status: Expired due to Fees

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Lindsay, Bruce G., Hamel, Elizabeth B.
• Primary Examiner(s): Flynn; Nathan
• Assistant Examiner(s): JEAN GILLES, JUDE
• Application Number:
  US10/143,317
• Publication Number:
  US 20030212789A1
• Time in Patent Office: 2,322 Days
• Field of Search: 709200-204, 709/225, 709/235, 709/245, 709/224, 709217-219, 709/220, 707201-204, 707 1- 10, 707/200, 707/103, 707/203, 710/20, 710/52, 370/230, 370/235, 370/352, 370/299, 714/15, 714/20, 714 5- 13, 711/122
• US Class Current: 709/225
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation

Abstract

Database management methods that exploit time dependent sequential database management system processes to ensure presentation of the same data or view to one or a plurality of users through sequencing asynchronous database management operations such as recovery and replication. Sequencing is accomplished through the use of entries in sequential logs, including transaction logs, recovery logs, and other data recovery tools and applications. Uses include managing data migration and data replication.

Claims

1. A method of managing a database management system that includes a replication system and that has at least one database table that is subject to replication, a recovery log, and a replication signal table, said method executing on a processor and in communication with a memory, the method comprising:

creating entries in the recovery log in response to changes to a state of any one or more of the at least one database table and the replication signal table, wherein said replication signal table is a specific table that is configured for all users of the database management system, and wherein the entries in the recovery log are ordered according to a total order;

inserting into the replication signal table, at least one row that describes a replication command to the replication system relating to the at least one database table;

detecting a recovery log entry in the recovery log that pertains to the at least one row in the replication signal table;

taking replication actions responsive to the recovery log entries that are positioned subsequent to the recovery log entry that pertains to the at least one row in the replication signal table, according to the total order of the recovery log; and

wherein the replication command comprises a first command comprising a signal stopping the replication of changes related to a database table wherein the first command is a signal stopping application of changes to data into a first database table, and a second command is a signal starting application of changes to contents of the first database table into a second database table.

2. The method of claim 1 wherein the replication command comprises a second command, and wherein the second command is a signal starting replication of changes to a database table.

3. The method of claim 2 wherein the database management system comprises at least two database tables, and the second command is a signal starting application of changes of contents of the first database table into second database table.

4. The method of claim 1 wherein the first command is a signal stopping capturing of changes to data into the first database table, and the second command is a signal capturing changes into the database table.

5. The method of claim 1 comprising performing an operation relating to any database table between the first command and the second command, and thereafter executing the second command.

6. The method of claim 5 comprising changing the schema or metadata of the database table between the first command and the second command, and thereafter executing the second command.

7. The method of claim 1 comprising executing said command entered in the replication signal table only after taking actions in response to all preceding entries of the recovery log.

8. The method of claim 1 wherein said command entered in said replication signal table also controls the processing of database events in a same transaction as said command.