Using invariants to validate applications states

• US 7,457,791 B1
• Filed: 05/30/2003
• Issued: 11/25/2008
• Est. Priority Date: 05/30/2003
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Rajasekaran, Kulothungan, Carter, Cale D., Martin, Deuane J., Singh, Pankaj K., Garg, Neeraj, Jayaram, Mahesh
• Primary Examiner(s): STARKS, WILBERT L
• Application Number:
  US10/449,208
• Time in Patent Office: 2,006 Days
• Field of Search: 706/45, 706/47, 707/5, 709/202, 709/230, 726/22, 235/472.01, 375/265, 715/716, 382/123, 700/83
• US Class Current: 706/47
• CPC Class Codes: G06F 11/0706 the processing taking place on a specific hardware platform or in a specific software environment
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/3604 Software analysis for verifying properties of programs byte-code verification G06F9/44589

Abstract

Architecture that facilitates discerning integrity of a data handling system. The architecture includes an identification component that periodically applies to data, in a batch mode, a context-based rule, the rule identifying anamolous data that does not satisfy the rule. A healing component is included to further rectify and/or flag the identified anomalous data.

Claims

1. A computer-implemented system that facilitates discerning integrity of a data handling system, comprising:

an identification component that periodically applies to data stored in a database accessed by an application, in a batch mode, a context-based rule, the rule is an invariant that is invoked at the application level to identify anomalous data that does not satisfy the rule, the anomalous data includes inconsistent data obtained during data migration; and

a healing component that at least one of rectifies or flags the identified anomalous data.

2. The system of claim 1, the rule including one or more conditions that are applied against the data, the results of which determine the integrity of the data handling system.

3. The system of claim 1, the rule applied to the data at least one of manually and automatically.

4. The system of claim 1, where application of the rule to the data returns a dataset that is a subset of the data.

5. The system of claim 4, the dataset is derived from at least one condition of the rule.

6. The system of claim 4, the dataset is derived from the application of multiple rules to the data.

7. The system of claim 4, the dataset derived from applying a first condition of the rule to the data followed by subsequent conditions of the rule until anomalous data is identified.

8. The system of claim 1, the rule is invoked at the application level during migration of the data from a first application version to second application version.

9. The system of claim 1, the rule is applied to a single database that is distributed across several locations.

10. The system of claim 1, the rule is processed into multiple subparts, each rule subpart applied against correspondingly separate sets of the data.

11. The system of claim 1, the rule is processed into multiple subparts where each rule subpart is applied against the data or a correspondingly separate subset of the data, each subset located on respective systems, or at least in combinations of subsets on the systems.

12. The system of claim 1, the rule is applied to at least one of a running system to ensure data integrity is maintained and to manage forks of program code over periods of time to ensure semantics of the data.

13. The system of claim 1, the rule is invoked from an application that is at a location remote from the data.

14. The system of claim 1, the rule invoked from an application that is at a location remote from the data, the rule communicated from the remote location to the data location for application against the data.

15. The system of claim 1, the rule invoked from an application that is at a location remote from the data, the rule communicated from the remote location to a corresponding application local to the data, for application against the data.

16. The system of claim 1, the rule automatically applied in response to a triggering event.

17. The system of claim 1, the identification component returning a dataset in response to applying the rule to the data, and if the dataset is too large, the rule is narrowed to return a smaller dataset.

18. The system of claim 17, the rule is narrowed at least one of automatically and manually.

19. The system of claim 17, the rule is narrowed by introducing additional conditions into the rule and applying the conditions of the rule successively to reduce the size of the returned dataset.

20. The system of claim 17, the data handling system automatically notified to expect a reduced dataset of the data with which to process its request.

21. The system of claim 1, the identification component returning a dataset in response to application of the rule to the data, and if the dataset is too large, one or more different rules are applied against the dataset to reduce the size of the dataset.

22. The system of claim 1, further comprising a support vector machine for prioritizing rule conditions to return a manageable dataset of the data.

23. The system of claim 1, further comprising a support vector machine for prioritizing a plurality of different rules that may be applied against the data to return a manageable dataset of the data.

24. A computer according to the system of claim 1.

25. A network having disposed thereon at least one computer operable according to the system of claim 1.

26. The system of claim 1, the identification component and healing component comprising a rule application that is at least one of integrated into the data handling system and accessible external to the data handling system.

27. A computer readable medium having stored thereon the components of claim 1.

28. A computer-implemented graphical user interface that facilitates management of a data integrity system for discerning the integrity of a data handling system, comprising:

an input component for receiving management information associated with periodic application of a context-based rule to data stored in a database that is accessed by an application, the context-based rule is an invariant specific to the application that is invoked at the application level; and

a presentation component for presenting a representation of the management information, the management information includes at least information regarding anomalous data identified by application of the context-based rule, the anomalous data includes inconsistent data obtained during data migration.

29. The interface of claim 28, the management information including one or more configuration inputs for selecting at least one of a data source, data destination, one or more rules to be applied against the data, data type, type of data handling, size of the returned dataset, and triggering event(s).

30. The interface of claim 28, further including a user workspace for receiving user input for creating a rule that is applied against the data.

31. The interface of claim 30, the user workspace for receiving a system status information.

32. The interface of claim 28 facilitating the connection to a remote site via at least one of a hyperlink and a TCP/IP communication link.

33. A computer-implemented method of discerning integrity of a data handling system, comprising:

periodically applying to data stored in a database that is accessed by an application, in a batch mode a context-based rule, the rule is an invariant specific to the application that is invoked at the application level to identify anomalous data that does not satisfy the rule, the anomalous data includes inconsistent data obtained during migration of data from a first application to a second application; and

correcting the anomalous data.

34. The method of claim 33, further comprising at least one of tagging anomalous data that cannot be corrected and transmitting a message in response thereto.

35. The method of claim 33, the rule is invoked automatically or manually at an application level.

36. The method of claim 33, where application of the rule to the data returns a dataset that is a subset of the data, the dataset is derived from at least one of a condition of the rule and multiple rules.

37. The method of claim 36, the dataset derived from applying a first condition of the rule to the data followed by subsequent conditions of the rule until anomalous data is identified.

38. The method of claim 33, the rule is invoked at the application level during migration of the data from a first application to second application.

39. The method of claim 33, the rule is applied to a single database that is distributed in parts across several locations.

40. The method of claim 33, the rule is processed into multiple subparts, each rule subpart applied against correspondingly separate sets of the data.

41. The method of claim 33, the rule applied periodically according to at least one of a predetermined time, a data migration, the amount of anomalous data, the type of anomalous data, the importance of the data, and the frequency of anomalous data.

42. The method of claim 33, the rule is processed into multiple subparts where each rule subpart is applied against the data or a correspondingly separate subset of the data, each subset located on respective systems, or at least in combinations of subsets on the systems.

43. The method of claim 33, the rule is invoked from at least one of an application local to the data and an application that is remote from the data.

44. The method of claim 33, the rule is invoked by at least one of passing the rule from a remote location to a local rule application for processing and transmitting a signal to the local rule application instructing the rule to be invoked.

45. The method of claim 33, the rule automatically applied in response to a triggering event.

46. The method of claim 33, further comprising automatically limiting the amount of returned data after applying the rule by at least one of applying a narrower rule, successively applying more conditions of the rule, applying more related conditions according to a prioritization algorithm, selectively applying one or more different rules according to a prioritization algorithm, and returning the anomalous data in parts.

47. The method of claim 46, the prioritization algorithm operating according to a support vector machine.

48. A computer-implemented system for discerning the integrity of a data handling system, comprising:

means for periodically applying to data stored in a database that is accessed by an application, in a batch mode, a context-based rule, the rule is an invariant specific to the application that is invoked at the application level to identify anomalous data that does not satisfy the rule, the anomalous data includes inconsistent data obtained during data migration;

means for correcting the anomalous data; and

means for tagging anomalous data that cannot be corrected.