Complex data query support in a partitioned database system

• US 6,405,198 B1
• Filed: 09/04/1998
• Issued: 06/11/2002
• Est. Priority Date: 09/04/1998
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Jang, Jing-Song, Bitar, Roger Georges, Kane, Erik Allan, Reyda, Angela Go, Ho, Jean Chen, Keesey, James Louis, Wilmot, Gerald Johann
• Primary Examiner(s): Coby, Frantz
• Application Number:
  US09/148,460
• Time in Patent Office: 1,376 Days
• Field of Search: 707/3, 707/4, 707/200, 707/5, 707/6
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2282 Tablespace storage structures; Management thereof
• 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 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

In a database processing system with a partitioned database that provides relational extenders to support storage of and access to complex data types, metadata table rows are collocated with corresponding relational table partitions. Image indexes are generated and stored at system nodes with their corresponding relational table partitions. Image query processing is conditioned to account for the partitioned structure of the database.

Claims

1. A computer-executed method for changing a relational table in a partitioned relational database system, the relational table having at least one column defined for a complex data type, the system including one or more metadata tables for extending a relation defined by the relational table, comprising:

dividing the relational table into partitions;

storing each partition at a respective node of the system; and

for each partition, storing rows of a metadata table that are associated with contents of the partition at the node where the partition is stored.

2. The method of claim 1, wherein the complex data type is an image data type.

3. A computer-executed method for entering information concerning a complex data type into a partitioned database of a relational database processing system, the database including one or more metadata tables for extending a relation of the database, comprising:

receiving a complex data object for storage in the database;

determining a location of a partition of the database in which the object is to be entered;

entering the object in the partition; and

storing relational extender metadata information for the object at the location.

4. The method of claim 3, wherein the complex data object is an image.

5. A computer-executed method for processing images in a partitioned relational database system, comprising:

receiving a request to insert an image into a partitioned table;

determining a node of the system where a partition of the table is stored;

storing an entry representing the image in the partition; and

at the node, generating an index representing the image and cataloging the index.

6. A computer-executed method for processing images in a relational database processing system with a partitioned database, comprising:

receiving an image to be stored in the partitioned database;

determining a database partition where the image is to be stored;

storing an entry representing the image in the partition; and

at the partition, generating an index representing the image and cataloging the index.

7. A computer-executed method for processing an image query in a relational database processing system with a partitioned database, comprising:

receiving a request to execute a query identifying an image stored in the partitioned database;

determining a database partition containing a reference to the image;

executing the query against all images referenced in the partition; and

returning a result of the query execution.

8. A computer-executed method for processing an image query in a database processing system with a partitioned database, comprising:

receiving a request to execute an image query identifying a table containing at least one image data type that is stored in the partitioned database;

determining nodes of the system where all partitions of the table are stored;

executing the query at each of the nodes;

receiving from each node results of the query execution;

merging the results from all of the nodes into a single result; and

returning the single result.

9. A relational database processing system for changing a relational table in a partitioned relational database system including at least one relational table having at least one column defined for a complex data type, the database system including one or more metadata tables for extending a relation defined by the relational table, comprising:

the relational table being divided into partitions;

each partition being stored at a respective node of the system; and

for each partition, rows of a metadata table that are associated with contents of the partition being stored at the node where the partition is stored.

10. The system of claim 9, wherein the complex data type is an image data type.

11. A relational database processing system for entering information concerning a complex data type into a partitioned database, the database including one or more metadata tables for extending a relation of the database, comprising:

at least one relational table with at least one column defining a complex data type;

the table being divided into partitions, each partition being stored at a particular system location; and

relational extender metadata information associated with each partition being stored at the location where the partition is stored.

12. The system of claim 11, wherein the complex data object is an image.

13. A partitioned database system for processing images, comprising:

a partitioned table having at least one column for defining an image data type;

each partition being stored at a node of the system; and

at each node, means for generating at least one index representing an image referenced in the partition stored at the node.

14. A database processing system with a partitioned database for processing images, comprising:

means for determining a database partition where an image is to be stored;

means for storing an entry representing the image in the partition; and

at the partition, means for generating an index representing the image and cataloging the index.

15. A database processing system with a partitioned database for processing a query identifying an image stored in the partitioned database, comprising:

means for determining a database partition containing a reference to the image;

means for executing the query against all images referenced in the partition; and

means for returning a result of the query execution.

16. A database processing system with a partitioned database for processing an image query identifying a table containing at least one image data type that is stored in the partitioned database, comprising:

means for determining nodes of the system where all partitions of the table are stored;

means for executing the query at each of the nodes;

means for receiving from each node results of the query execution;

means for merging the results from all of the nodes into a single result; and

means for returning the single result.

17. Computer software for changing a relational table in a partitioned relational database system, the relational table having at least one column defined for a complex data type, the system including one or more metadata tables for extending a relation defined by the relational table, the computer executable code comprising:

software executable for dividing the relational table into partitions;

software executable for storing each partition at a respective node of the system; and

software executable for storing rows of a metadata table for each partition, the rows being associated with contents of the partition at the node where the partition is stored.

18. The computer executable code of claim 17, wherein the complex data type is an image data type.

19. Computer software for processing an image query in a database processing system with a partitioned database, the computer executable code comprising:

software executable to receive a request to execute an image query identifying a table containing at least one image data type that is stored in the partitioned database;

software executable to determine nodes of the system where all partitions of the table are stored;

software executable to execute the query at each of the nodes;

software executable to receive from each node results of the query execution;

software executable to merge the results from all of the nodes into a single result; and

software executable to return the single result.

20. A computer readable code means on a computer usable medium for changing a relational table in a partitioned relational database system, the relational table having at least one column defined for a complex data type, the system including one or more metadata tables for extending a relation defined by the relational table, the computer readable code means comprising:

dividing the relational table into partitions;

storing each partition at a respective node of the systems; and

for each partition, storing rows of a metadata table that are associated with contents of the partition at the node where the partition is stored.

21. The computer readable code means of claim 20, wherein the complex data type is an image data type.

22. A computer readable code means on a computer usable medium for entering information concerning a complex data type into a partitioned database of a relational database processing system, the database including one or more metadata tables for extending a relation of the database, the computer readable code means comprising:

receiving a complex data object for storage in the database;

determining a location of a partition of the database in which the object is to be entered;

entering the object in the partition; and

storing relational extender metadata information for the object at the location.

23. The computer readable code means of claim 22, wherein the complex data object is an image.

24. A computer readable code means on a computer usable medium for processing images in a partitioned relational database system, the computer readable code means comprising:

receiving a request to insert an image into a partitioned table;

determining a node of the system where a partition of the table is stored;

storing an entry representing the image in the partition; and

at the node, generating an index representing the image and cataloging the index.

25. A computer readable code means on a computer usable medium for processing images in a relational database processing system with a partitioned database, the computer code means comprising:

receiving an image to be stored in the partitioned database;

determining a database partition where the image is to be stored;

storing an entry representing the image in the partition; and

at the partition, generating an index representing the image and cataloging the index.

26. A computer readable code means on a computer usable medium for processing an image query in a relational database processing system with a partitioned database, the computer readable code means comprising:

receiving an request to execute a query identifying an image stored in the partitioned database;

determining a database partition containing a reference to the image;

executing the query against all images referenced in the partition; and

returning a result of the query execution.

27. A computer readable code means on a computer usable medium for processing an image query in a database processing system with a partitioned database, the computer readable code means comprising:

receiving a request to execute an image query identifying a table containing at least one image data type that is stored in the partitioned database;

determining nodes of the system where all partitions of the table are stored;

executing the query at each of the nodes;

receiving from each node results of the query execution;

merging the results from all of the nodes into a single result; and

returning the single result.