METHOD OF PROCESSING RELATIONAL QUERIES IN A DATABASE SYSTEM AND CORRESPONDING DATABASE SYSTEM

• US 20140012882A1
• Filed: 07/11/2012
• Published: 01/09/2014
• Est. Priority Date: 07/04/2012
• Status: Active Grant

• Current Assignee: Software AG
• Original Assignee: Software AG
• Inventors: POPPITZ, Michael
• Granted Patent: US 9,400,815 B2
• Time in Patent Office: Days
• US Class Current: 707/792
• CPC Class Codes: G06F 16/2228 Indexing structures
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/2456 Join operations
• CPC Class Codes: G06F 16/24561 Intermediate data storage techniques for performance improvement

Abstract

Certain example embodiments concern a computer-implemented method of processing a relational query in a database system. The relational query addresses a plurality of data objects linked by one or more relationships. For each data object addressed by the relational query, at least one result is computed in accordance with at least one index structure, with the index structure being stored in a storage device of the database system. The results computed are merged in accordance with at least one translation data structure to produce a final result of the relational query, with the translation data structure representing the one or more relationships between the data objects and being stored in a memory of the database system.

Claims

1. A computer-implemented method of processing a relational query in a database system, wherein the relational query addresses a plurality of data objects linked by one or more relationships, method comprising:

a. for each data object addressed by the relational query, computing, in connection with a processor, at least one result in accordance with at least one index structure, the index structure being stored in a storage device of the database system; and

b. merging, in connection with a or the processor, the results computed in (a) in accordance with at least one translation data structure to produce a final result of the relational query, the translation data structure representing the one or more relationships between the data objects and being stored in a memory of the database system.

2. The method of claim 1, further comprising representing the relational query as a query tree, wherein each data object addressed by the relational query is represented as a node of the query tree and wherein each relationship between the data objects is represented by an edge in the query tree.

3. The method of claim 2, further comprising performing the computing and/or the merging as part of a first pass of processing the query tree, wherein the first pass is performed recursively for each node of the query tree.

4. The method of the preceding claim 3, wherein the merging comprises merging the results of the current node with the results of the child node(s) of the current node.

5. The method of claim 2, further comprising performing a second merging as part of a second pass of processing the query tree, wherein the second pass is performed recursively for each node of the query tree.

6. The method of claim 5, wherein the second merging comprises merging the results of the current node with the results of the parent node of the current node.

7. The method of claim 3, wherein the first pass and/or the second pass of processing the query tree traverses the query tree starting at the root node in a depth-first manner.

8. The method of claim 1, wherein for two given data objects which are addressed by the relational query and which are linked by a relationship, the translation data structure comprises a mapping of an identifier of the first data object to an identifier of the second data object in accordance with the relationship.

9. The method of claim 8, wherein, if the relationship is represented as a data object in a database of the database system, the translation data structure comprises a mapping of an identifier of the first data object to an identifier of the second data object and to an identifier of the relationship.

10. A non-transitory computer readable storage medium tangibly storing a computer program comprising instructions for implementing a method of claim 1.

11. A database system, comprising:

a. an interface configured to receive relational queries which address a plurality of data objects linked by one or more relationships;

b. a storage device configured to store at least one index structure;

c. a memory configured to store at least one translation data structure representing the one or more relationships between the data objects;

d. control logic that, in connection with at least one processor, is configured to at least;

for each data object addressed by the relational query, compute at least one result in accordance with the at least one index structure; and

merge the computed results in accordance with the at least one translation data structure to produce a final result of the relational query.

12. The database system of claim 11, wherein the control logic is further configured to represent the relational query as a query tree, wherein each data object addressed by the relational query is represented as a node of the query tree and wherein each relationship between the data objects is represented by an edge in the query tree.

13. The database system of claim 12, wherein the control logic is further configured to perform the computing and/or the merging as part of a first pass of processing the query tree, wherein the first pass is performed recursively for each node of the query tree.

14. The database system of claim 12, wherein the control logic is further configured to perform a second merging as part of a second pass of processing the query tree, wherein the second pass is performed recursively for each node of the query tree.

15. The database system of claim 13, wherein the first pass and/or the second pass of processing the query tree traverses the query tree starting at the root node in a depth-first manner.