Apparatus and method for coordinating logical data replication with highly available data replication

• US 20040158588A1
• Filed: 02/07/2003
• Published: 08/12/2004
• Est. Priority Date: 02/07/2003
• Status: Active Grant

• Current Assignee: Daedalus Blue LLC
• Original Assignee: International Business Machines Corporation
• Inventors: Pruet, Clarence Madison III
• Granted Patent: US 7,177,886 B2
• Time in Patent Office: Days
• US Class Current: 707/204
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

In a database apparatus (10), a critical database server (12) includes a primary server (20) supporting a primary database instance and a secondary server (22) supporting a secondary database instance that mirrors the primary database instance. The secondary server (22) generates an acknowledgment signal (60) indicating that a selected critical database transaction (42) is mirrored at the secondary database instance. A plurality of other servers (14, 16, 18) each support a database. A data replicator (30) communicates with the critical database server (12) and the other servers (14, 16, 18) to replicate the selected critical database transaction (42) on at least one of said plurality of other servers (14, 16, 18) responsive to the acknowledgment signal (60).

Claims

1. A database apparatus including:

a critical database server including a primary server supporting a primary database instance and a secondary server supporting a secondary database instance that mirrors the primary database instance, the secondary server generating an acknowledgment signal indicating that a selected critical database transaction is mirrored at the secondary database instance;

a plurality of other servers each supporting a database; and

a data replicator communicating with the critical database server and the other servers to replicate the selected critical database transaction on at least one of said plurality of other servers responsive to the acknowledgment signal.

2. The database apparatus as set forth in claim 1, wherein the critical database server includes:

a mirroring component communicating with the primary and secondary servers to transfer database log file entries of the primary database instance to the secondary server, the secondary server applying and logging the transferred database log file entries to the secondary database instance and producing said acknowledgement signal subsequent to the applying and logging of the selected critical database transaction.

3. The database apparatus as set forth in claim 2, wherein the mirroring component includes:

a control structure that indexes critical database transactions that are applied and logged at the secondary database instance, the acknowledgement signal corresponding to indexing in the control structure of at least one of the selected critical database transaction and a critical database transaction that commits after the selected critical database transaction.

4. The database apparatus as set forth in claim 1, wherein the primary and secondary servers of the critical database server are remotely located from one another, the database apparatus further including:

a communication network that communicates data between the primary and secondary servers and between the critical database server and the other servers.

5. The database apparatus as set forth in claim 1, wherein the data replicator includes:

a send queue that stores one or more transactions that effect replication of the selected critical database transaction.

6. The database apparatus as set forth in claim 5, wherein the selected critical database transaction is received by the critical database server from the data replicator, the data replicator further including:

a replication acknowledgment queue that buffers a replication acknowledgment sent by the critical database server to the data replicator; and

a mirror monitor that transmits the queued replication acknowledgment to one of the other servers responsive to the acknowledgment signal.

7. A method for integrating a high availability replication system that produces at least one mirror of a critical database node, and a data distribution replication system that selectively replicates data at least from the critical database node to one or more remote database nodes, the method including:

in the data distribution replication system, identifying an object at the critical database node targeted for replication;

in the high availability replication system, replicating objects including the identified object at the mirror and generating a mirror acknowledgment indicative of completion of replication of the identified object at the mirror; and

in the data distribution replication system, replicating the identified object responsive to the mirror acknowledgment.

8. The method as set forth in claim 7, wherein the generating of a mirror acknowledgment includes:

recording log positions of objects replicated at the mirror; and

generating the mirror acknowledgment responsive to a recorded log position exceeding a log position of the identified object.

9. The method as set forth in claim 7, wherein the generating of an acknowledgment includes:

in the data distribution replication system, replicating the identified object to the critical database node; and

in the data distribution replication system, communicating a data distribution replication acknowledgment from the critical database node responsive to the mirror acknowledgment.

10. A method for coordinating data replication to distributed database servers with a hot-backup instance of a database, the method including:

backing up database transactions at the hot-backup instance;

maintaining a backup indicator that identifies database transactions backed up at the hot-backup source; and

delaying data replication of a database transaction until the backup indicator identifies the database transaction as having been backed up at the hot-backup source.

11. The method as set forth in claim 10, wherein the backing up of database transactions at the hot-backup instance includes:

copying database logs to the hot-backup instance.

12. The method as set forth in claim 10, wherein the database transaction is received from a replication thread that replicates a transaction at a selected one of the distributed database servers, the method further including:

delaying sending a replication acknowledgment to the replication thread until the backup indicator identifies the database transaction as having been backed up at the hot-backup source.

13. An article of manufacture including a program storage medium readable by a computer and embodying one or more instructions executable by the computer to perform process operations for executing a command to perform a database operation on a relational database connected to the computer, the method including:

identifying a transaction performed in the relational database; and

replicating the identified transaction responsive to an indication that the identified transaction has been backed up at the relational database.

14. The article of manufacture as set forth in claim 13, wherein the relational database includes a primary database in which the identified transaction has performed, and a secondary database, and the method further includes:

copying a record of the identified transaction from the primary database to the secondary database to effect the backing up of the identified transaction at the relational database.

15. The article of manufacture as set forth in claim 14, wherein the method further includes:

subsequent to the copying, communicating an acknowledgment from the secondary database indicating that the copying has been performed, the communicated acknowledgment effecting the indication that the identified transaction has been backed up at the relational database.

16. The article of manufacture as set forth in claim 14, wherein the method further includes:

recording information on transaction records copied from the primary server to the secondary server, the indication that the identified transaction has been backed up at the relational database being generated based on the recorded information.

17. The article of manufacture as set forth in claim 13, wherein the identified transaction is a replication of a transaction performed at a remote database server, and the method further includes:

communicating an acknowledgment that the identified transaction has been processed in the relational database, the communicating being responsive to the indication that the identified transaction has been backed up at the relational database.

18. Apparatus for supporting an associated distributed relational database including a primary server supporting a primary database instance with a primary database instance log file, a secondary server supporting a secondary database instance with a secondary instance log file, and a plurality of other servers each supporting a database instance, the apparatus including:

a highly available data replication component communicating with the primary and secondary servers to transfer primary database instance log file entries from the primary server to the secondary server, the secondary server producing an acknowledgment indicating that the transferred log file entries have been received; and

a logical data replication component communicating with the primary server and the other servers to identify a log record in the primary database instance log file, construct a replication transaction corresponding to the identified log record, and, responsive to the highly available data replication component indicating that the identified log record has been received at the secondary server, cause one or more of the other servers to perform the replication transaction.

19. The apparatus as set forth in claim 18, wherein the logical data replication component produces computation threads for replicating transactions, the apparatus further including:

a thread-verification component that terminates a computation thread if it is communicating with the secondary server.

20. The apparatus as set forth in claim 18, further including:

a critical node database server logical entity that communicates with the logical data replication component as a single logical entity, the critical node database server logical entity representing the primary and secondary servers.

21. The apparatus as set forth in claim 18, wherein the highly available data replication component includes:

an acknowledgment log index that identifies a position in the primary database instance log file corresponding to a most recent primary database instance log file entry transferred from the primary server to the secondary server and acknowledged by the secondary server, the indication that the identified log record has been received at the secondary server corresponding to the acknowledgment log index identifying a position in the primary database instance log file that is more recent than the identified log record.

22. The apparatus as set forth in claim 18, wherein the logical data replication component produces computation threads for replicating transactions, and the logical data replication component includes:

a thread terminator which terminates a computation thread responsive to a replication acknowledgment indicating a replication transaction generated by the computation thread has been performed by one or more servers.

23. The apparatus as set forth in claim 22, wherein the logical data replication component further includes:

a replication acknowledgment buffer that stores a replication acknowledgment received from the primary server indicating the primary server has performed the replication transaction generated by the computation thread, the replication acknowledgment buffer transmitting the replication acknowledgment to the computation thread responsive to the highly available data replication component indicating that the replication transaction generated by the computation thread has been received at the secondary server.

24. The apparatus as set forth in claim 18, wherein the distributed relational database further includes a second primary server supporting a second primary database instance with a second primary database instance log file, and a second secondary server supporting a second secondary database instance with a second secondary database instance log file, the apparatus further including:

a second highly available data replication component communicating with the second primary and second secondary servers to transfer second primary database instance log file entries from the second primary server to the second secondary server, the second secondary server producing a second acknowledgment indicating that the transferred log file entries have been received;

wherein the logical data replication component additionally communicates with the second primary server and the second secondary server to identify a second log record in the second primary database instance log file, construct a second replication transaction corresponding to the identified second log record, and, responsive to the second highly available data replication component indicating that the identified second log record has been received at the second secondary server, cause at least one of the other servers and the primary server to perform the second replication transaction.