Method, system and program for executing a query having a UNION operator

• US 20060101001A1
• Filed: 11/05/2004
• Published: 05/11/2006
• Est. Priority Date: 11/05/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Zuzarte, Calisto Paul, Pirahesh, Mir Hamid, Sharpe, David C., Lindsay, Bruce Gilbert, Liu, Linqi, Sutyanyong, Nattavut, Neugebauer, Robert Paul
• Granted Patent: US 7,409,385 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24537 of operators
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

Disclosed is a data processing system implemented method, a data processing system and an article of manufacture for executing a query having a union operator. The data processing system implemented method directs the data processing system to process a query against data objects. The data objects are operatively coupled to the data processing system. The query includes a parent operator. The parent operator references a union operator. The union operator references sub-queries. The sub-queries reference the data objects. The data processing system implemented method includes noting a set of partitionings for the union operator, the noted set of partitionings being based on the sub-queries and being based on the data objects reference by the sub-queries, and executing the query having the union operator, the execution of the query being based on the noted set of partitionings and the parent operator.

Claims

1. A data processing system implemented method of directing a data processing system to process a query against data objects, the data objects being operatively coupled to the data processing system, the query including a parent operator, the parent operator referencing a union operator, the union operator referencing sub-queries, the sub-queries referencing the data objects, the data processing system implemented method comprising:

noting a set of partitionings for the union operator, the noted set of partitionings being based on the sub-queries and being based on the data objects reference by the sub-queries; and

executing the query having the union operator, the execution of the query being based on the noted set of partitionings and the parent operator.

2. The data processing system implemented method of claim 1 wherein the union operator is a union all operator.

3. The data processing system implemented method of claim 1 wherein the noting of the set of partitionings for the union operator comprises:

gathering data restrictions for the data objects referenced by the union operator;

analyzing the gathered data restrictions, the analyzed gathered data restrictions identifying valid data contained in the data objects; and

generating the set of partitionings based on the analyzed gathered data.

4. The data processing system implemented method of claim 3 wherein the gathering of the data restrictions comprises:

retrieving, for each data object, data constraints, functional dependencies and keys from a catalog of the data processing system;

retrieving predicates and expressions from a sub-query selected from the sub-queries, the retrieved predicates and the retrieved expressions placing additional restrictions on data contained in the data objects; and

inferring additional data restrictions from a combination of the retrieved data constraints, the retrieved functional dependencies, the retrieved keys, the retrieved predicates and the retrieved expressions.

5. The data processing system implemented method of claim 4 wherein the inferring of the additional data restrictions comprises:

demarcating a value graph according to the retrieved data constraints, the retrieved predicates and the retrieved expressions; and

adding additional demarcations by combining the retrieved functional dependencies and the retrieved keys with existing demarcations contained in the value graph.

6. The data processing system implemented method of claim 3 wherein the analyzing of the gathered data restrictions comprises:

mapping dimensions of value graphs for individual data objects to dimensions of the union operator; and

combining the value graphs and the mapped dimensions into a value graph for the union operator with additional annotations associated with regions contained in the value graph, the annotations produced by the data objects.

7. The data processing system implemented method of claim 6 wherein the mapping of the dimensions of the value graphs comprises:

determining which dimension of each data object that corresponds to each particular dimension associated with the union operator by following a data flow from the union operator down to the each data object; and

storing a set of corresponding dimensions in the mapped dimensions.

8. The data processing system implemented method of claim 6 wherein the combining of the value graphs comprises:

adding demarcations associated with the value graph of each data object to the value graph associated with the union operator; and

adding an additional annotation to regions associated with the added demarcations, the additional annotation indicating which data object the demarcation originated therefrom.

9. The data processing system implemented method of claim 3 wherein the generating of the set of partitioning comprises:

examining all combinations of the dimensions in the value graph associated with the union operator to determine the minimal combination of dimensions that can uniquely identify each region in the value graph, the demarcations and the annotations of the value graph related to the minimal combination is a partitioning.

10. The data processing system implemented method of claim 1 wherein the executing of the query having the union operator comprises:

matching predicates and functional dependencies associated with the parent operator to each partitioning;

discarding any unsatisfactory partitionings;

determining an optimal partitioning using a combination of a costing method, a statistical method and a heuristic method; and

using the determined optimal partitioning, the matched predicates and the matched functional dependencies for the purpose of executing the query against the data objects, the executing query accessing those data objects that are relevant to the query.

11. The data processing system implemented method of claim 10 wherein the matching of the each partitioning comprises:

finding the predicates and the functional dependencies that involve any dimension of the partitioning; and

discarding the partitioning if any dimension does not have at least one of a suitable predicate and a suitable functional dependency, when every dimension of the partitioning has at least one suitable predicate then the partitioning is satisfied.

12. The data processing system implemented method of claim 10 wherein the using of the determined optimal partitioning comprises:

determining which data objects provides data relevant to the query by combining the matched predicates, the matched functional dependencies and the optimal partitioning; and

retrieving data from the determined data objects.

13. A data processing system for processing a query against data objects, the data objects being operatively coupled to the data processing system, the query including a parent operator, the parent operator referencing a union operator, the union operator referencing sub-queries, the sub-queries referencing the data objects, the data processing system comprising:

a noting module for noting a set of partitionings for the union operator, the noted set of partitionings being based on the sub-queries and being based on the data objects reference by the sub-queries; and

an executing module for executing the query having the union operator, the execution of the query being based on the noted set of partitionings and the parent operator.

14. The data processing system of claim 13 wherein the union operator is a union all operator.

15. The data processing system of claim 13 wherein the noting module for noting of the set of partitionings for the union operator comprises:

a gathering module for gathering data restrictions for the data objects referenced by the union operator;

an analyzing module for analyzing the gathered data restrictions, the analyzed gathered data restrictions identifying valid data contained in the data objects; and

a generating module for generating the set of partitionings based on the analyzed gathered data.

16. The data processing system of claim 15 wherein the gathering module for gathering of the data restrictions comprises:

a first retrieving module for retrieving, for each data object, data constraints, functional dependencies and keys from a catalog of the data processing system;

a second retrieving module for retrieving predicates and expressions from a sub-query selected from the sub-queries, the retrieved predicates and the retrieved expressions placing additional restrictions on data contained in the data objects; and

an inferring module for inferring additional data restrictions from a combination of the retrieved data constraints, the retrieved functional dependencies, the retrieved keys, the retrieved predicates and the retrieved expressions.

17. The data processing system of claim 16 wherein the inferring module for inferring of the additional data restrictions comprises:

a demarcating module for demarcating a value graph according to the retrieved data constraints, the retrieved predicates and the retrieved expressions; and

an adding module for adding additional demarcations by combining the retrieved functional dependencies and the retrieved keys with existing demarcations contained in the value graph.

18. The data processing system of claim 15 wherein the analyzing module for analyzing of the gathered data restrictions comprises:

a mapping module for mapping dimensions of value graphs for individual data objects to dimensions of the union operator; and

a combining module for combining the value graphs and the mapped dimensions into a value graph for the union operator with additional annotations associated with regions contained in the value graph, the annotations produced by the data objects.

19. The data processing system of claim 18 wherein the mapping module for mapping of the dimensions of the value graphs comprises:

a determination module for determining which dimension of each data object that corresponds to each particular dimension associated with the union operator by following a data flow from the union operator down to the each data object; and

a storing module for storing a set of corresponding dimensions in the mapped dimensions.

20. The data processing system of claim 18 wherein the combining of the value graphs comprises:

an addition module for adding demarcations associated with the value graph of each data object to the value graph associated with the union operator; and

an adding module for adding an additional annotation to regions associated with the added demarcations, the additional annotation indicating which data object the demarcation originated therefrom.

21. A data processing system of claim 15 wherein the generating module for generating of the set of partitioning comprises:

an examining module for examining all combinations of the dimensions in the value graph associated with the union operator to determine the minimal combination of dimensions that can uniquely identify each region in the value graph, the demarcations and the annotations of the value graph related to the minimal combination is a partitioning.

22. The data processing system of claim 13 wherein the executing module for executing of the query having the union operator comprises:

a matching module for matching predicates and for matching functional dependencies associated with the parent operator to each partitioning;

a discarding module for discarding any unsatisfactory partitionings;

a determining module for determining an optimal partitioning using a combination of a costing method, a statistical method and a heuristic method; and

a using module for using the determined optimal partitioning, the matched predicates and the matched functional dependencies for the purpose of executing the query against the data objects, the executing query accessing those data objects that are relevant to the query.

23. A data processing system of claim 22 wherein the matching module for matching of the each partitioning comprises:

a finding module for finding the predicates and the functional dependencies that involve any dimension of the partitioning;

a discarding module for discarding the partitioning if any dimension does not have at least one of a suitable predicate and a suitable functional dependency, when every dimension of the partitioning has at least one suitable predicate then the partitioning is satisfied.

24. The data processing system of claim 22 wherein the using module for using of the determined optimal partitioning comprises:

a determination module for determining which data objects provides data relevant to the query by combining the matched predicates, the matched functional dependencies and the optimal partitioning; and

a retrieving module for retrieving data from the determined data objects.

25. An article of manufacture for directing a data processing system to process a query against data objects, the data objects being operatively coupled to the data processing system, the query including a parent operator, the parent operator referencing a union operator, the union operator referencing sub-queries, the sub-queries referencing the data objects, the article of manufacture comprising:

a program usable medium embodying one or more instructions executable by the data processing system, the one or more instructions comprising;

data processing system executable instructions for noting a set of partitionings for the union operator, the noted set of partitionings being based on the sub-queries and being based on the data objects reference by the sub-queries; and

data processing system executable instructions for executing the query having the union operator, the execution of the query being based on the noted set of partitionings and the parent operator.

26. The article of manufacture of claim 25 wherein the union operator is a union all operator.

27. The article of manufacture of claim 25 wherein the data processing system executable instructions for noting of the set of partitionings for the union operator comprises:

data processing system executable instructions for gathering data restrictions for the data objects referenced by the union operator;

data processing system executable instructions for analyzing the gathered data restrictions, the analyzed gathered data restrictions identifying valid data contained in the data objects; and

data processing system executable instructions for generating the set of partitionings based on the analyzed gathered data.

28. The article of manufacture of claim 27 wherein the data processing system executable instructions for gathering of the data restrictions comprises:

data processing system executable instructions for retrieving, for each data object, data constraints, functional dependencies and keys from a catalog of the data processing system;

data processing system executable instructions for retrieving predicates and expressions from a sub-query selected from the sub-queries, the retrieved predicates and the retrieved expressions placing additional restrictions on data contained in the data objects; and

data processing system executable instructions for inferring additional data restrictions from a combination of the retrieved data constraints, the retrieved functional dependencies, the retrieved keys, the retrieved predicates and the retrieved expressions.

29. The article of manufacture of claim 28 wherein the data processing system executable instructions for inferring of the additional data restrictions comprises:

data processing system executable instructions for demarcating a value graph according to the retrieved data constraints, the retrieved predicates and the retrieved expressions; and

data processing system executable instructions for adding additional demarcations by combining the retrieved functional dependencies and the retrieved keys with existing demarcations contained in the value graph.

30. The article of manufacture of claim 27 wherein the data processing system executable instructions for analyzing of the gathered data restrictions comprises:

data processing system executable instructions for mapping dimensions of value graphs for individual data objects to dimensions of the union operator; and

data processing system executable instructions for combining the value graphs and the mapped dimensions into a value graph for the union operator with additional annotations associated with regions contained in the value graph, the annotations produced by the data objects.

31. The article of manufacture of claim 30 where the data processing system executable instructions for mapping of the dimensions of the value graphs comprises:

data processing system executable instructions for determining which dimension of each data object that corresponds to each particular dimension associated with the union operator by following a data flow from the union operator down to the each data object; and

data processing system executable instructions for storing a set of corresponding dimensions in the mapped dimensions.

32. The article of manufacture of claim 30 wherein the data processing system executable instructions for combining of the value graphs comprises:

data processing system executable instructions for adding demarcations associated with the value graph of each data object to the value graph associated with the union operator; and

data processing system executable instructions for adding an additional annotation to regions associated with the added demarcations, the additional annotation indicating which data object the demarcation originated therefrom.

33. The article of manufacture of claim 27 wherein the data processing system executable instructions for generating of the set of partitioning comprises:

data processing system executable instructions for examining all combinations of the dimensions in the value graph associated with the union operator to determine the minimal combination of dimensions that can uniquely identify each region in the value graph, the demarcations and the annotations of the value graph related to the minimal combination is a partitioning.

34. The article of manufacture of claim 25 wherein the data processing system executable instructions for executing of the query having the union operator comprises:

data processing system executable instructions for matching predicates and functional dependencies associated with the parent operator to each partitioning;

data processing system executable instructions for discarding any unsatisfactory partitionings;

data processing system executable instructions for determining an optimal partitioning using a combination of a costing method, a statistical method and a heuristic method; and

data processing system executable instructions for using the determined optimal partitioning, the matched predicates and the matched functional dependencies for the purpose of executing the query against the data objects, the executing query accessing those data objects that are relevant to the query.

35. The article of manufacture of claim 34 wherein the data processing system executable instructions for matching of the each partitioning comprises:

data processing system executable instructions for finding the predicates and the functional dependencies that involve any dimension of the partitioning; and

data processing system executable instructions for discarding the partitioning if any dimension does not have at least one of a suitable predicate and a suitable functional dependency, when every dimension of the partitioning has at least one suitable predicate then the partitioning is satisfied.

36. The article of manufacture of claim 34 wherein the data processing system executable instructions for using of the determined optimal partitioning comprises:

data processing system executable instructions for determining which data objects provides data relevant to the query by combining the matched predicates, the matched functional dependencies and the optimal partitioning; and

data processing system executable instructions for retrieving data from the determined data objects.