Application-independent data synchronization technique

• US 6,505,200 B1
• Filed: 07/06/2000
• Issued: 01/07/2003
• Est. Priority Date: 07/06/2000
• Status: Expired due to Term

• Current Assignee: King.com, Ltd (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Li, Yongcheng, Ims, Steven D., Tan, Yih-Shin
• Primary Examiner(s): Coby, Frantz
• Application Number:
  US09/611,030
• Time in Patent Office: 915 Days
• Field of Search: 707/8, 707/203, 707/204, 707/3, 707/2, 707/4
• US Class Current: 707/638
• CPC Class Codes: G06F 16/1794 Details of file format conversion
• 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: Y10S 707/955 Object-oriented
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99938 Concurrency, e.g. lock management in shared database

Abstract

A method, system, and computer program product for automatically synchronizing data between a replicated version and a back-end data store version which may or may not have the same format. Queues are used for scheduling refreshes of read-access objects and updates of write-access objects, where the actual processing will occur when the client device connects to the back-end data source. This technique is application-independent, and does not require writing application-specific code to perform a synchronization between data stores having dissimilar storage formats. This technique is particularly advantageous for mobile computing devices, which may operate primarily in a disconnected mode, and for distributed branch-office computing environments where a local server has a replicated copy of data stored at a back-end enterprise system. This technique is also advantageous where the back-end data source uses legacy host application or database storage formats, and where the client software uses modem object-oriented data storage formats. A technique is defined for offloading the processing to a remote machine.

Claims

1. A computer program product for synchronizing replicated data, the computer program product embodied on one or more computer-readable media and comprising:

computer-readable program code means for storing one or more first objects as replicated read-access objects in a first cache for responding to read requests against the first objects, wherein (1) a set of input properties and values thereof is stored with or associated with each replicated read-access object and (2) refresh logic specifying how to refresh each of the replicated read-access objects is stored with or associated with the replicated read-access object or a group of replicated read-access objects;

computer-readable program code means for storing one or more second objects as replicated write-access objects in a second cache for responding to update requests against the it second objects, wherein (1) a set of input properties is stored with or associated with each replicated write-access object and (2) update logic specifying how to update each of the replicated write-access objects is stored with or associated with the replicated write-access object or a group of replicated write-access objects;

computer-readable program code means for receiving read requests against one or more of the first objects;

computer-readable program code means for receiving update requests against one or more of the second objects;

computer-readable program code means for responding to the read requests using the replicated read-access objects;

computer-readable program code means for queuing the update requests, along with the input properties and values thereof which are to be used for performing each update request, as queued update requests on an update queue;

computer-readable program code means for scheduling a refresh of a selected replicated read-access object by queuing the selected replicated read-access object or a reference thereto as a queued refresh request on a refresh queue;

computer-readable program code means for determining that a replication is to be performed; and

computer-readable program code means for performing the replication to refresh the replicated read-access objects and the replicated write-access objects by processing the queued refresh requests on the refresh queue and the queued update requests on the update queue.

2. The computer program product according to claim 1, wherein a separate refresh queue and a separate update queue is created for each of one or more back-end data sources to be accessed during operation of the computer-readable program code means for performing the replication.

3. The computer program product according to claim 1, wherein the computer-readable program code means for performing the replication further comprises:

computer-readable program code means for executing the refresh logic stored with or associated with selected replicated read-access objects for which the queued refresh requests are queued; and

computer-readable program code means for executing the update logic stored with or associated with selected replication write-access objects for which the queued update requests are queued.

4. The computer program product according to claim 1, wherein a first caching policy of the refresh queue is set to refresh the replicated read-access objects upon making a connection to a first back-end data source and a second caching policy of the update queue is set to perform the queued update requests on the update queue upon making the connection to the first back-end data source or upon making a connection to a second back-end data source.

5. The computer program product according to claim 4, wherein operation of the computer-readable program code means for performing the replication is triggered according to one or both of the first caching policy or the second caching policy.

6. The computer program product according to claim 1, further comprising:

computer-readable program code means for connecting to one or more back-end data sources prior to operation of the computer-readable program code means for performing the replication; and

computer-readable program code means for disconnecting from the one or more back-end data sources after operation of the computer-readable program code means for performing the replication.

7. The computer program product according to claim 1, wherein the computer-readable program code means for performing the replication further comprises computer-readable program code means for offloading the replication to a remote device.

8. The computer program product according to claim 7, wherein the computer-readable program code means for offloading the replication to the remote device further comprises:

computer-readable program code means for packaging the refresh queue and update queue for sending to the remote device;

computer-readable program code means for sending the packaged queues to the remote device;

computer-readable program code means for receiving a response from the remote device which indicates that the offloaded replication has been performed;

computer-readable program code means for refreshing the replicated read-access objects, responsive to operation of the computer-readable program code means for receiving the response; and

computer-readable program code means for purging the refresh queue and update queue, responsive to operation of the computer-readable program code means for refreshing.

9. The computer program product according to claim 8, wherein the received response comprises information to use during operation of the computer-readable program code means for refreshing.

10. The computer program product according to claim 9, wherein the received response further comprises an Extensible Markup Language (XML) representation of the information to use.

11. The computer program product according to claim 8, wherein the computer-readable program code means for packaging further comprises computer-readable program code means for creating an Extensible Markup Language (XML) representation of the refresh queue and the update queue.

12. The computer program product according to claim 8, further comprising:

computer-readable program code means for connecting to the remote device prior to operation of the computer-readable program code means for sending the packaged queues; and

computer-readable program code means for disconnecting from the remote device after operation of the computer-readable program code means for receiving the response.

13. The computer program product according to claim 1, wherein the computer-readable program code means for performing the replication further comprises:

computer-readable program code means for processing the queued update requests on the update queue, further comprising;

computer-readable program code means for setting the input properties of a selected replicated write-access object against which the queued update request is to be performed using the queued input property values; and

computer-readable program code means for executing the update logic stored with or associated with the selected replicated write-access object using the input properties and values thereof; and

computer-readable program code means for processing the queued refresh requests on the refresh queue, after operation of the computer-readable program code means for processing the queued update requests, by executing the refresh logic stored with or associated with selected replicated read-access objects for which the queued refresh requests are queued, thereby refreshing the selected replicated read-access objects.

14. The computer program product according to claim 13, further comprising:

computer-readable program code means for connecting to one or more back-end data sources prior to operation of the computer-readable program code means for processing the queued update requests; and

computer-readable program code means for disconnecting from the one or more back-end data sources after operation of the computer-readable program code means for processing the queued refresh requests.

15. The computer program product according to claim 1, wherein the computer-readable program code means for determining that the replication is to be performed comprises computer-readable program code means for detecting that a connection to a back-end data source has been made.

16. A system for synchronizing replicated data, comprising:

means for storing one or more first objects as replicated read-access objects in a first cache for responding to read requests against the first objects, wherein (1) a set of input properties and values thereof is stored with or associated with each replicated read-access object and (2) refresh logic specifying how to refresh each of the replicated read-access objects is stored with or associated with the replicated read-access object or a group of replicated read-access objects;

means for storing one or more second objects as replicated write-access objects in a second cache for responding to update requests against the second objects, wherein (1) a set of input properties is stored with or associated with each replicated write-access object and (2) update logic specifying how to update each of the replicated write-access objects is stored with or associated with the replicated write-access object or a group of replicated write-access objects;

means for receiving read requests against one or more of the first objects;

means for receiving update requests against one or more of the second objects;

means for responding to the read requests using the replicated read-access objects;

means for queuing the update requests, along with the input properties and values thereof which are to be used for performing each update request, as queued update requests on an update queue;

means for scheduling a refresh of a selected replicated read-access object by queuing the selected replicated read-access object or a reference thereto as a queued refresh request on a refresh queue;

means for determining that a replication is to be performed; and

means for performing the replication to refresh the replicated read-access objects and the replicated write-access objects by processing the queued refresh requests on the refresh queue and the queued update requests on the update queue.

17. The system according to claim 16, wherein a separate refresh queue and a separate update queue is created for each of one or more back-end data sources to be accessed during operation of the means for performing the replication.

18. The system according to claim 16, wherein the means for performing the replication further comprises:

means for executing the refresh logic stored with or associated with selected replicated read-access objects for which the queued refresh requests are queued; and

means for executing the update logic stored with or associated with selected replication write-access objects for which the queued update requests are queued.

19. The system according to claim 16, wherein a first caching policy of the refresh queue is set to refresh the replicated read-access objects upon making a connection to a first back-end data source and a second caching policy of the update queue is set to perform the queued update requests on the update queue upon making the connection to the first back-end data source or upon making a connection to a second back-end data source.

20. The system according to claim 19, wherein operation of the means for performing the replication is triggered according to one or both of the first caching policy or the second caching policy.

21. The system according to claim 16, further comprising:

means for connecting to one or more back-end data sources prior to operation of the means for performing the replication; and

means for disconnecting from the one or more back-end data sources after operation of the means for performing the replication.

22. The system according to claim 16, wherein the means for performing the replication further comprises means for offloading the replication to a remote device.

23. The system according to claim 22, wherein the means for offloading the replication to the remote device further comprises:

means for packaging the refresh queue and update queue for sending to the remote device;

means for sending the packaged queues to the remote device;

means for receiving a response from the remote device which indicates that the offloaded replication has been performed;

means for refreshing the replicated read-access objects, responsive to operation of the means for receiving the response; and

means for purging the refresh queue and update queue, responsive to operation of the means for refreshing.

24. The system according to claim 23, wherein the received response comprises information to use during operation of the means for refreshing.

25. The system according to claim 24, wherein the received response further comprises an Extensible Markup Language (XML) representation of the information to use.

26. The system according to claim 23, wherein the means for packaging further comprises means for creating an Extensible Markup Language (XML) representation of the refresh queue and the update queue.

27. The system according to claim 23, further comprising:

means for connecting to the remote device prior to operation of the means for sending the packaged queues; and

means for disconnecting from the remote device after operation of the means for receiving the response.

28. The system according to claim 16, wherein the means for performing the replication further comprises:

means for processing the queued update requests on the update queue, further comprising;

means for setting the input properties of a selected replicated write-access object against which the queued update request is to be performed using the queued input property values; and

means for executing the update logic stored with or associated with the selected replicated write-access object using the input properties and values thereof; and

means for processing the queued refresh requests on the refresh queue, after operation of the means for processing the queued update requests, by executing the refresh logic stored with or associated with selected replicated read-access objects for which the queued refresh requests are queued, thereby refreshing the selected replicated read-access objects.

29. The system according to claim 28, further comprising:

means for connecting to one or more back-end data sources prior to operation of the means for processing the queued update requests; and

means for disconnecting from the one or more back-end data sources after operation of the means for processing the queued refresh requests.

30. The system according to claim 16, wherein the means for determining that the replication is to be performed comprises means for detecting that a connection to a back-end data source has been made.

31. A method performed by one or more computers for synchronizing replicated data, comprising the steps of:

storing one or more first objects as replicated read-access objects in a first cache for responding to read requests against the first objects, wherein (1) a set of input properties and values thereof is stored with or associated with each replicated read-access object and (2) refresh logic specifying how to refresh each of the replicated read-access objects is stored with or associated with the replicated read-access object or a group of replicated read-access objects;

storing one or more second objects as replicated write-access objects in a second cache for responding to update requests against the second objects, wherein (1) a set of input properties is stored with or associated with each replicated write-access object and (2) update logic specifying how to update each of the replicated write-access objects is stored with or associated with the replicated write-access object or a group of replicated write-access objects;

receiving read requests against one or more of the first objects;

receiving update requests against one or more of the second objects;

responding to the read requests using the replicated read-access objects;

queuing the update requests, along with the input properties and values thereof which are to be used for performing each update request, as queued update requests on an update queue;

scheduling a refresh of a selected replicated read-access object by queuing the selected replicated read-access object or a reference thereto as a queued refresh request on a refresh queue;

determining that a replication is to be performed; and

performing the replication to refresh the replicated read-access objects and the replicated write-access objects by processing the queued refresh requests on the refresh queue and the queued update requests on the update queue.

32. The method according to claim 31, wherein a separate refresh queue and a separate update queue is created for each of one or more back-end data sources to be accessed during the step of performing the replication.

33. The method according to claim 31, wherein the step of performing the replication further comprises the steps of:

executing the refresh logic stored with or associated with selected replicated read-access objects for which the queued refresh requests are queued; and

executing the update logic stored with or associated with selected replication write-access objects for which the queued update requests are queued.

34. The method according to claim 31, wherein a first caching policy of the refresh queue is set to refresh the replicated read-access objects upon making a connection to a first back-end data source and a second caching policy of the update queue is set to perform the queued update requests on the update queue upon making the connection to the first back-end data source or upon making a connection to a second back-end data source.

35. The method according to claim 34, wherein the step of performing the replication is triggered according to one or both of the first caching policy or the second caching policy.

36. The method according to claim 31, further comprising the steps of:

connecting to one or more back-end data sources prior to performing the replication; and

disconnecting from the one or more back-end data sources after performing the replication.

37. The method according to claim 31, wherein the step of performing the replication further comprises the step of offloading the replication to a remote device.

38. The method according to claim 37, wherein the step of offloading the replication to the remote device further comprises the steps of:

packaging the refresh queue and update queue for sending to the remote device;

sending the packaged queues to the remote device;

receiving a response from the remote device which indicates that the offloaded replication has been performed;

refreshing the replicated read-access objects, responsive to the step of receiving the response; and

purging the refresh queue and update queue, responsive to the step of refreshing.

39. The method according to claim 38, wherein the received response comprises information to use during the refreshing step.

40. The method according to claim 39, wherein the received response further comprises an Extensible Markup Language (XML) representation of the information to use.

41. The method according to claim 38, wherein the packaging step further comprises the step of creating an Extensible Markup Language (XML) representation of the refresh queue and the update queue.

42. The method according to claim 38, further comprising the steps of

connecting to the remote device prior to sending the packaged queues;

and

disconnecting from the remote device after receiving the response.

43. The method according to claim 31, wherein the step of performing the replication further comprises the steps of:

processing the queued update requests on the update queue, further comprising the steps of;

setting the input properties of a selected replicated write-access object against which the queued update request is to be performed using the queued input property values; and

executing the update logic stored with or associated with the selected replicated write-access object using the input properties and values thereof; and

processing the queued refresh requests on the refresh queue, after processing the queued update requests, by executing the refresh logic stored with or associated with selected replicated read-access objects for which the queued refresh requests are queued, thereby refreshing the selected replicated read-access objects.

44. The method according to claim 43, further comprising the steps of:

connecting to one or more back-end data sources prior to processing the queued update requests; and

disconnecting from the one or more back-end data sources after processing the queued refresh requests.

45. The method according to claim 31, wherein the step of determining that the replication is to be performed comprises the step of detecting that a connection to a back-end data source has been made.