System and method for adaptive query identification and acceleration

• US 20060155697A1
• Filed: 12/23/2004
• Published: 07/13/2006
• Est. Priority Date: 12/23/2004
• Status: Active Grant

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Rosengard, Phillip I.
• Granted Patent: US 7,596,560 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2255 Hash tables
• CPC Class Codes: G06F 16/24539 using cached or materialised query results
• CPC Class Codes: G06F 16/2454 Optimisation of common expressions
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

A data management system and method. The system is adapted to analyze a plurality of data queries and provide a metric with respect thereto. The system then orders the queries based on the metric to improve a performance parameter of the system or estimate an execution time with respect to the query. In a more specific implementation, the system includes code for analyzing a signature of each of the queries, comparing the signature to a calibrated signature and providing a signature score in response thereto. The system further includes code for estimating an execution time of at least one of the queries. The signature may be a cluster of statements in an appropriate language such as SQL (standard query language). As an alternative, the signature may be a set of features extracted from a plurality of statements.

Claims

1. A data management system comprising:

first means for analyzing a plurality of data queries and providing a metric with respect thereto and

second means for ordering said queries based on said metric to improve a performance parameter of said system.

2. The invention of claim 1 wherein said first means includes means for analyzing a signature of each of said queries.

3. The invention of claim 2 wherein said signature is a cluster of statements.

4. The invention of claim 3 wherein said statements are standard query language (SQL) statements.

5. The invention of claim 2 wherein said signature is a set of features extracted from said a plurality of statements.

6. The invention of claim 5 wherein said statements are standard query language (SQL) statements.

7. The invention of claim 2 wherein said first means includes means for comparing said signature to a calibrated signature and providing a signature score in response thereto.

8. The invention of claim 7 further including means responsive to said score for estimating an execution time of at least one of said queries.

9. A data management system comprising:

first means for analyzing plurality of data queries and

second means responsive to said first means for estimating an execution time of a query.

10. The invention of claim 9 wherein said first means includes means for analyzing a signature of each of said queries.

11. The invention of claim 10 wherein said signature is a cluster of statements.

12. The invention of claim 11 wherein said statements are standard query language (SQL) statements.

13. The invention of claim 10 wherein said signature is a set of features extracted from said a plurality of statements.

14. The invention of claim 13 wherein said statements are standard query language (SQL) statements.

15. The invention of claim 10 wherein said first means includes means for comparing said signature to a calibrated signature and providing a signature score in response thereto.

16. The invention of claim 15 further including means responsive to said score for ordering said queries to minimize a metric with respect thereto.

17. The invention of claim 16 wherein said metric is execution time.

18. A data management system comprising:

a database;

a processor coupled to said database;

a communications interface for supplying a plurality of queries to said database via said processor; and

software running on said processor for analyzing a plurality of said queries and organizing said queries to optimize a performance parameter with respect to a retrieval thereof.

19. The invention of claim 18 wherein said software includes code for analyzing a signature of each of said queries.

20. The invention of claim 19 wherein said signature is a cluster of statements.

21. The invention of claim 20 wherein said statements are standard query language (SQL) statements.

22. The invention of claim 19 wherein said signature is a set of features extracted from said a plurality of statements.

23. The invention of claim 22 wherein said statements are standard query language (SQL) statements.

24. The invention of claim 19 wherein said software includes code for comparing said signature to a calibrated signature and providing a signature score in response thereto.

25. The invention of claim 24 further including code responsive to said score for ordering said queries to minimize a metric with respect thereto.

26. The invention of claim 25 wherein said metric is execution time.

27. A data management method including the steps of:

analyzing a plurality of data queries and providing a metric with respect thereto and

ordering said queries based on said metric to improve a performance parameter thereof.