Automated abstract database generation through existing application statement analysis

• US 7,143,081 B2
• Filed: 02/12/2003
• Issued: 11/28/2006
• Est. Priority Date: 02/12/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Dettinger, Richard D., Stevens, Richard J., Will, Eric W., Kulack, Frederick A.
• Primary Examiner(s): Alam, Hosain
• Assistant Examiner(s): Saeed, Usmaan
• Application Number:
  US10/365,297
• Publication Number:
  US 20040158556A1
• Time in Patent Office: 1,385 Days
• Field of Search: 707 1- 10, 707/100, 707/202, 707/203
• US Class Current: 1/1
• CPC Class Codes: G06F 16/217 Database tuning G06F16/2282 takes precedence; database performance monitoring G06F11/3409
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

The present invention generally is directed to methods, articles of manufacture and systems for automatically generating abstract representations of a limited subset of data repository fields accessed by a particular application. The limited subset of fields may be identified by monitoring queries issued against the data repository by the application. The abstract representations may be grouped in a data repository abstraction (DRA) component that serves as a data definition framework allowing data from the fields to be accessed independent of the particular manner in which the data is physically represented in the data repository. By limiting the fields exposed to the application to the limited subset of fields actually accessed by the application, the DRA component, in effect, provides the application with a custom database tailored to its particular needs.

Claims

1. A computer implemented method for automatically generating a data repository abstraction component describing, and used to access, data in a data repository, the method comprising:

monitoring queries issued against the data repository by an application;

collecting, from the monitored queries, information regarding usage of fields of the data repository;

identifying, from the collected usage information, a limited subset of one or more fields of the data repository accessed by the application; and

generating, for use in the data repository abstraction component, abstract representations for the limited subset of fields;

wherein the abstract representations logically model the respective limited subset of fields;

whereby the data repository abstraction component is customized on the basis of usage of the fields of the data repository;

wherein the generated abstract representations model the respective fields of the limited subset of fields and expose the data to users in a manner that allows the users to compose queries on the basis of the generated abstract representations without knowledge of or regard for a particular schema of the data repository;

wherein each of the abstract representations comprises;

a logical field definition comprising a logical field name;

at least one location attribute identifying a location of data in the data repository corresponding to the logical field name; and

a reference to an access method invoked to access the data in response to receiving an abstract query containing the logical field name.

2. The method of claim 1, wherein identifying the limited subset of fields comprises extracting, from the monitored queries, a list of one or more referenced fields.

3. The method of claim 2, further comprising determining, based on the monitored queries, whether one or more of the referenced fields is a searchable field.

4. The method of claim 2, further comprising determining, based on the monitored queries, whether one or more of the referenced fields is returned as query results.

5. The method of claim 2, further comprising determining, from the monitored queries, a logical name for the at least one referenced field to be included in a corresponding abstract representation.

6. The method of claim 2, further extracting Information regarding a relationship between at least two referenced fields.

7. The method of claim 1, wherein monitoring the queries comprises examining an output modifier used in a monitored query.

8. The method of claim 1, further comprising gathering metadata for results returned for the monitored queries.

9. The method of claim 8, further comprising obtaining, from the metadata, a list of fields returned as query results.

10. The method of claim 9, wherein generating the abstract representations comprises generating an abstract representation for at least one of the fields returned as query results.

11. A computer implemented method for automatically generating abstract representations of one or more fields of a data source, the method comprising:

monitoring queries issued against the data source;

collecting, from the monitored queries, information regarding usage of the fields of the data source; and

based on the collected usage information, generating the abstract representations for select fields of the data source, wherein the generated abstract representations model the respective select fields and expose the data to users in a manner that allows the users to compose queries on the basis of the generated abstract representations without knowledge of or regard fur a particular schema of the data source;

wherein each of the abstract representations comprises a logical field definition comprising a logical field name, at least one location attribute identifying a location of physical data corresponding to the logical field name, and a reference to an access method invoked to access the physical data from the data source in response to receiving an abstract query containing the logical field name.

12. The method of claim 11, wherein collecting information regarding usage of the fields comprises extracting a list of one or more fields of the data source referenced by the monitored queries.

13. The method of claim 12, wherein generating the abstract representations comprises generating an abstract representation for at least one of the fields referenced by the monitored queries.

14. The method of claim 13, wherein the abstract representation for the at least one field comprises an indication of a data type for the at least one field, the data type determined from the collected information regarding usage of the fields.

15. The method of claim 12, further comprising, determining a relationship between at least two of the fields referenced by the monitored queries, based on the collected information regarding usage of the fields.

16. The method of claim 11, wherein at least one of the abstract representations comprises a logical name for a corresponding field of the data source, the logical name determined from the collected information regarding usage of the fields.

17. A tangible computer readable storage medium containing a program which, when executed, performs operations for automatically generating a data repository abstraction component describing, and used to access, data in a data repository, the operations comprising:

monitoring queries issued against the data repository;

collecting usage information from the monitored queries, the usage information pertaining to how the fields of the data source are being used; and

generating, based on the collected usage information, a set of logical field specifications for a limited subset of one or more fields of the data repository accessed by the monitored queries for use in the data repository abstraction component;

wherein the generated logical field specifications model the respective select fields and expose the data to users in a manner that allows the users to compose abstract queries on the basis of the data repository abstraction component without knowledge of or regard for a particular schema of the data repository;

wherein each of the logical field specifications comprises a logical field definition comprising a logical field name, at least one location attribute identifying a location of physical data corresponding to the logical field name, and a reference to an access method invoked to access the physical data from the data source in response to receiving an abstract query containing the logical field name.

18. The tangible computer readable storage medium of claim 17, wherein collecting usage information from the monitored queries comprises extracting a list of fields referenced by the monitored queries.

19. The tangible computer readable storage medium of claim 18, wherein the operations further comprise determining, based on the collected usage information, whether at least one of the referenced fields is a searchable field.

20. The tangible computer readable storage medium of claim 18, wherein the operations further comprise determining, from the collected usage information, a logical name for at least one of the referenced fields to be included in a corresponding logical field specification.

21. The tangible computer readable storage medium of claim 17, wherein:

the data repository is a relational database;

collecting usage information comprises extracting a list of tables and corresponding columns referenced by the monitored queries; and

generating the set of logical field specifications comprises generating a logical field specification for at least one of the referenced columns.

22. The tangible computer readable storage medium of claim 17, wherein collecting usage information comprises gathering metadata for results returned for the monitored queries.

23. The tangible computer readable storage medium of claim 22, wherein collecting usage information comprises obtaining, from the metadata, a list of fields returned as results.

24. The tangible computer readable storage medium of claim 23, wherein generating the logical field specifications comprises generating a logical field specification for at least one of the fields returned as results.

25. A data processing system, comprising:

a processor;

a data repository containing data;

an application suitable for querying the data repository; and

a data repository abstraction generator configured to generate a data repository abstraction component describing, and used to access, data in the data repository based on usage information collected by monitoring and parsing queries issued against the data repository by the application;

wherein the data repository abstraction generator is configured to generate a plurality of logical field specifications for a limited subset of fields of the data repository accessed by the application, as indicated by the monitored queries;

whereby the data repository abstraction is customized on the basis of how the application queries the data repository;

wherein the generated logical field specifications model the respective fields of the limited subset of fields and expose the physical data to users in a manner that allows the users to compose queries on the basis of the generated logical field specifications without knowledge of or regard for a particular schema of the data repository;

wherein each of the logical field specifications comprises;

a logical field definition comprising a logical field name;

at least one location attribute identifying a location of data in the data repository corresponding to the logical field name; and

a reference to an access method invoked to access the data in response to receiving an abstract query containing the logical field name.

26. The data processing system of claim 25, wherein the data repository abstraction generator is configured to determine a logical name, to include in one or more of the generated logical field specifications, based on reference to one or more corresponding fields in the data repository in the monitored queries.

27. The data processing system of claim 25, wherein the data repository comprises a relational database and the data repository generator is configured to extract a list of referenced tables and associated columns from the monitored queries and generate logical field specifications for at least one of the columns.

28. The data processing system of claim 25, wherein the collected usage information comprises output modifiers within the modified queries.

29. The data processing system of claim 25, wherein the data repository abstraction generator is configured to gather metadata for results returned for the monitored queries.

30. The data processing system of claim 29, wherein the data repository abstraction generator is further configured to obtain, from the metadata, a list of fields and associated data types returned as results for the monitored queries.

31. The data processing system of claim 30, wherein the data repository abstraction generator is configured to generate a logical field specification including the associated data type for one or more of the fields returned as results for the monitored queries.