Buffered message queue architecture for database management systems with guaranteed at least once delivery

• US 7,185,034 B2
• Filed: 05/21/2003
• Issued: 02/27/2007
• Est. Priority Date: 08/01/2002
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Meiyyappan, Krishnan, Mishra, Shailendra, Surlaker, Kapil, Jain, Namit, Nainani, Bhagat, Bhatt, Neerja
• Primary Examiner(s): Coby; Frantz
• Application Number:
  US10/443,323
• Publication Number:
  US 20040034640A1
• Time in Patent Office: 1,378 Days
• Field of Search: 707/102, 707/203, 707/100, 707/101, 707/103.R, 707/104.1, 707/200, 707/206, 707/1, 719/314, 714/2, 709/206, 709/207
• US Class Current: 1/1
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: Y10S 707/99957 Garbage collection

Abstract

A buffered message queue architecture for managing messages in a database management system is disclosed. A “buffered message queue” refers to a message queue implemented in a volatile memory, such as a RAM. The volatile memory may be a shared volatile memory that is accessible by a plurality of processes. The buffered message queue architecture supports a publish and subscribe communication mechanism, where the message producers and message consumers may be decoupled from and independent of each other. The buffered message queue architecture provides all the functionality of a persistent publish-subscriber messaging system, without ever having to store the messages in persistent storage. The buffered message queue architecture provides better performance and scalability since no persistent operations are needed and no UNDO/REDO logs need to be maintained. Messages published to the buffered message queue are delivered to all eligible subscribers at least once, even in the event of failures, as long as the application is “repeatable.” The buffered message queue architecture also includes management mechanisms for performing buffered message queue cleanup and also for providing unlimited size buffered message queues when limited amounts of shared memory are available. The architecture also includes “zero copy” buffered message queues and provides for transaction-based enqueue of messages.

Claims

1. A method for managing messages in a database system, the method comprising the computer-implemented step of in response to a first message in a message queue created in a volatile memory of the database system being provided to all subscriber processes for which the first message satisfies the corresponding subscription data,

generating a checkpoint value that indicates that the first message does not need to be restored to the message queue after a failure that causes the loss of the first message from the message queue;

and

storing the checkpoint value on a non-volatile storage.

2. The method as recited in claim 1, further comprising the computer-implemented step of after a failure that causes the loss of the first message from the message queue,

retrieving the checkpoint value from the non-volatile storage, and

providing the checkpoint value to a publisher process and requesting that the publisher process re-publish to the message queue any messages that the publisher process previously published to the message queue prior to the failure and after the first message was published to the message queue.

3. The method as recited in claim 1, further comprising the computer-implemented step of in response to a second message in the message queue being provided to all subscriber processes for which the second message satisfies the corresponding subscription data, updating the checkpoint value on the non-volatile storage to indicate that the second message does not need to be restored to the message queue after a failure that causes the loss of the second message from the message queue.

4. A method for managing messages in a database system, the method comprising the computer-implemented steps of:

generating and storing on a non-volatile storage in the database system, a first checkpoint value that identifies a first most recent message that was enqueued by a first publisher into a message queue implemented in a volatile memory of the database system and dequeued by all subscribers to the message queue for which the first most recent message satisfies corresponding subscription data; and

generating and storing on the non-volatile storage a second checkpoint value that identifies a second most recent message that was enqueued by a second publisher into the message queue and dequeued by all subscribers to the message queue for which the second most recent message satisfies the corresponding subscription data.

5. The method as recited in claim 4, further comprising the computer-implemented step of after a failure,

retrieving the first checkpoint value from the non-volatile storage, and

providing the first checkpoint value to the first publisher so that the first publisher knows which messages the first publisher must be re-enqueue into the message queue.

6. The method as recited in claim 4, further comprising the computer-implemented step of in response to a third most recent message that was enqueued by the first publisher into the message queue being dequeued by all subscribers to the message queue for which the third most recent message satisfies the corresponding subscription data, updating the first checkpoint value on the non-volatile storage to identify the third most recent message.

7. A method for managing messages in a distributed database environment, the method comprising the computer-implemented steps of:

propagating one or more messages from a first message queue disposed in a volatile memory of a first database system to a second message queue disposed in a volatile memory of a second database system; and

receiving and storing in a non-volatile memory at the first database system, propagation acknowledgement data that indicates a most recent message from the one or more messages that have been processed for all subscribers to the second message queue of the second database system.

8. The method as recited in claim 7, further comprising the computer-implemented step of after a failure that causes the loss of any of the one or more messages in the second message queue of the second database system, re-propagating at least one message from the one or more messages stored in the first message queue in the first database system to the second message queue in the second database system, wherein the at least one message was enqueued in the first message queue of the first database system after the most recent message indicated by the propagation data.

9. The method as recited in claim 7, further comprising the computer-implemented steps of:

propagating one or more additional messages from the first message queue of the first database system to the second message queue of the second database system; and

receiving and storing in the non-volatile memory of the first database system second propagation acknowledgement data that indicates a most recent message from the one or more additional messages that have been processed for all subscribers to the second message queue of the second database system.

10. The method as recited in claim 7, further comprising the computer-implemented steps of at the first database system:

retrieving a checkpoint value that indicates a most recent message provided to all subscriber processes to the first message queue for which the first message satisfies the corresponding subscription data;

selecting as a selected message either the most recent message indicated by the checkpoint value or the message indicated by the propagation value based upon whether the most recent message indicated by the checkpoint value or the message indicated by the propagation value was stored earliest in the first message queue; and

deleting, from the first message queue of the first database system, one or more messages stored in the first message queue prior to the selected message.

11. A computer-readable storage medium for managing messages in a database system, the computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, cause the one or more processors to perform the step in response to a first message in a message queue created in a volatile memory of the database system being provided to all subscriber processes for which the first message satisfies the corresponding subscription data,

generating a checkpoint value that indicates that the first message does not need to be restored to the message queue after a failure that causes the loss of the first message from the message queue;

and

storing the checkpoint value on a non-volatile storage.

12. The computer-readable storage medium as recited in claim 11, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of after a failure that causes the loss of the first message from the message queue,

retrieving the checkpoint value from the non-volatile storage, and

providing the checkpoint value to a publisher process and requesting that the publisher process re-publish to the message queue any messages that the publisher process previously published to the message queue prior to the failure and after the first message was published to the message queue.

13. The computer-readable storage medium as recited in claim 11, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of in response to a second message in the message queue being provided to all subscriber processes for which the second message satisfies the corresponding subscription data, updating the checkpoint value on the non-volatile storage to indicate that the second message does not need to be restored to the message queue after a failure that causes the loss of the second message from the message queue.

14. A computer-readable storage medium for managing messages in a database system, the computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of:

generating and storing on a non-volatile storage in the database system, a first checkpoint value that identifies a first most recent message that was enqueued by a first publisher into a message queue implemented in a volatile memory of the database system and dequeued by all subscribers to the message queue for which the first most recent message satisfies corresponding subscription data; and

generating and storing on the non-volatile storage a second checkpoint value that identifies a second most recent message that was enqueued by a second publisher into the message queue and dequeued by all subscribers to the message queue for which the second most recent message satisfies the corresponding subscription data.

15. The computer-readable storage medium as recited in claim 14, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of after a failure,

retrieving the first checkpoint value from the non-volatile storage, and

providing the first checkpoint value to the first publisher so that the first publisher knows which messages the first publisher must be re-enqueue into the message queue.

16. The computer-readable storage medium as recited in claim 14, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of in response to a third most recent message that was enqueued by the first publisher into the message queue being dequeued by all subscribers to the message queue for which the third most recent message satisfies the corresponding subscription data, updating the first checkpoint value on the non-volatile storage to identify the third most recent message.

17. A computer-readable storage medium for managing messages in a distributed database environment, the computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of:

propagating one or more messages from a first message queue disposed in a volatile memory of a first database system to a second message queue disposed in a volatile memory of a second database system; and

receiving and storing in a non-volatile memory at the first database system, propagation acknowledgement data that indicates a most recent message from the one or more messages that have been processed for all subscribers to the second message queue of the second database system.

18. The computer-readable storage medium as recited in claim 17, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of after a failure that causes the loss of any of the one or more messages in the second message queue of the second database system, re-propagating at least one message from the one or more messages stored in the first message queue in the first database system to the second message queue in the second database system, wherein the at least one message was enqueued in the first message queue of the first database system after the most recent message indicated by the propagation data.

19. The computer-readable storage medium as recited in claim 17, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

propagating one or more additional messages from the first message queue of the first database system to the second message queue of the second database system; and

receiving and storing in the non-volatile memory of the first database system second propagation acknowledgement data that indicates a most recent message from the one or more additional messages that have been processed for all subscribers to the second message queue of the second database system.

20. The computer-readable storage medium as recited in claim 17, further comprising one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of at the first database system:

retrieving a checkpoint value that indicates a most recent message provided to all subscriber processes to the first message queue for which the first message satisfies the corresponding subscription data;

selecting as a selected message either the most recent message indicated by the checkpoint value or the message indicated by the propagation value based upon whether the most recent message indicated by the checkpoint value or the message indicated by the propagation value was stored earliest in the first message queue; and

deleting, from the first message queue of the first database system, one or more messages stored in the first message queue prior to the selected message.

21. An apparatus for managing messages in a database system, the apparatus comprising a memory carrying one or more sequences of instructions which, when executed by one or more processors, cause the one or more processors to perform the step in response to a first message in a message queue created in a volatile memory of the database system being provided to all subscriber processes for which the first message satisfies the corresponding subscription data,

generating a checkpoint value that indicates that the first message does not need to be restored to the message queue after a failure that causes the loss of the first message from the message queue;

and

storing the checkpoint value on a non-volatile storage.

22. The apparatus as recited in claim 21, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of after a failure that causes the loss of the first message from the message queue,

retrieving the checkpoint value from the non-volatile storage, and

providing the checkpoint value to a publisher process and requesting that the publisher process re-publish to the message queue any messages that the publisher process previously published to the message queue prior to the failure and after the first message was published to the message queue.

23. The apparatus as recited in claim 21, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of in response to a second message in the message queue being provided to all subscriber processes for which the second message satisfies the corresponding subscription data, updating the checkpoint value on the non-volatile storage to indicate that the second message does not need to be restored to the message queue after a failure that causes the loss of the second message from the message queue.

24. An apparatus for managing messages in a database system, the apparatus comprising a memory carrying one or more sequences of instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of:

generating and storing on a non-volatile storage in the database system, a first checkpoint value that identifies a first most recent message that was enqueued by a first publisher into a message queue implemented in a volatile memory of the database system and dequeued by all subscribers to the message queue for which the first most recent message satisfies corresponding subscription data; and

generating and storing on the non-volatile storage a second checkpoint value that identifies a second most recent message that was enqueued by a second publisher into the message queue and dequeued by all subscribers to the message queue for which the second most recent message satisfies the corresponding subscription data.

25. The apparatus as recited in claim 24, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, retrieving the first checkpoint value from the non-volatile storage, and

providing the first checkpoint value to the first publisher so that the first publisher knows which messages the first publisher must be re-enqueue into the message queue.

26. The apparatus as recited in claim 24, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of in response to a third most recent message that was enqueued by the first publisher into the message queue being dequeued by all subscribers to the message queue for which the third most recent message satisfies the corresponding subscription data, updating the first checkpoint value on the non-volatile storage to identify the third most recent message.

27. An apparatus for managing messages in a distributed database environment, the apparatus comprising a memory carrying one or more sequences of instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of:

propagating one or more messages from a first message queue disposed in a volatile memory of a first database system to a second message queue disposed in a volatile memory of a second database system; and

receiving and storing in a non-volatile memory at the first database system, propagation acknowledgement data that indicates a most recent message from the one or more messages that have been processed for all subscribers to the second message queue of the second database system.

28. The apparatus as recited in claim 27, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the step of after a failure that causes the loss of any of the one or more messages in the second message queue of the second database system, re-propagating at least one message from the one or more messages stored in the first message queue in the first database system to the second message queue in the second database system, wherein the at least one message was enqueued in the first message queue of the first database system after the most recent message indicated by the propagation data.

29. The apparatus as recited in claim 27, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

propagating one or more additional messages from the first message queue of the first database system to the second message queue of the second database system; and

receiving and storing in the non-volatile memory of the first database system second propagation acknowledgement data that indicates a most recent message from the one or more additional messages that have been processed for all subscribers to the second message queue of the second database system.

30. The apparatus as recited in claim 27, wherein the memory further includes one or more additional instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of at the first database system:

retrieving a checkpoint value that indicates a most recent message provided to all subscriber processes to the first message queue for which the first message satisfies the corresponding subscription data;

selecting as a selected message either the most recent message indicated by the checkpoint value or the message indicated by the propagation value based upon whether the most recent message indicated by the checkpoint value or the message indicated by the propagation value was stored earliest in the first message queue; and

deleting, from the first message queue of the first database system, one or more messages stored in the first message queue prior to the selected message.