Methods and apparatus for accesssing non-relational data files using relational queries

• US 5,379,419 A
• Filed: 12/07/1990
• Issued: 01/03/1995
• Est. Priority Date: 12/07/1990
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Digital Equipment Corporation (HP Inc.)
• Inventors: Sunkara, Ramu V., Pittman, Steven J., Heffernan, John S., Savage, Peter L.
• Primary Examiner(s): Kulik, Paul IV.
• Assistant Examiner(s): Loomis, John C.
• Application Number:
  US07/623,762
• Time in Patent Office: 1,488 Days
• Field of Search: 395/600, 364/DIG. 1:DIG. 2
• US Class Current: 1/1
• CPC Class Codes: G06F 16/258 Data format conversion from or to a database
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

A relational data access facility allows relation-type queries to access data stored in non-relational data files by converting relational queries into a set of common commands which are sent to data drivers to obtain the data specified by those queries. The facility uses metadata which describes the organization of the data in the non-relational files, and examines the expressions in the relational queries to formulate an access plan for the data. The plan is formulated to reduce cost and promote efficiency.

Claims

1. A method for providing relational access to data stored in non-relational data files in response to queries for accessing a relational database, the relational database of the type composed of tables comprising the steps, executed by a data processor, of:

maintaining metadata corresponding to each of the non-relational data files, the metadata defining the data organization in a corresponding non-relational data file;

accessing the metadata from the dictionary means;

decoding the queries, according to the metadata, into a subset of data access commands, the subset of data access commands selected from a set of data access commands common to the non-relational files; and

accessing the data in the non-relational data files specified by each of the subset of data access commands received from the data engine means according to the metadata.

2. The method of claim 1 further including the step of communicating with a dictionary driver to obtain information regarding the metadata for decoding the queries into the subset of data access commands.

3. The method of claim 1 further including the steps of ensuring the correctness of the queries;

and

forming Boolean expressions from the queries.

4. The method of claim 1 further including the step of placing the subset of data access commands into a sequence which expedites the access of the desired data.

5. The method of claim 4 wherein the step of placing the subset of data access commands into a sequence which expedites the access of data includes the substep of

organizing the Boolean expressions into a tree structure.

6. The method of claim 5 wherein the step of placing the subset of data access commands into a sequence which expedites the access of data includes the substep of

traversing the tree structure to form an access plan containing the subset of data access commands to access the desired data.

7. The method of claim 6 wherein the step of placing the subset of data access commands into a sequence which expedites the access of data includes the substep of

examining the amount of system resources used by different sequences of the subset of data access commands.

8. The method of claim 6 wherein the step of placing the subset of data access commands into a sequence which expedites the access of data includes the substep of

executing the subset of data access commands in the access plan.

9. A system for providing relational access to data stored in non-relational data files in response to queries for accessing a relational database, the relational database of the type composed of relationship tables, comprising:

dictionary means for maintaining metadata corresponding to each of the non-relational data files, the metadata defining a data organization in a corresponding non-relational data files;

data engine means, coupled to the dictionary means and isolated from the physical structure of the non-relational data files, for accessing the metadata from the dictionary means and for decoding the queries, according to the metadata, into a subset of data access commands, the subset of data access commands selected from a set of data access commands common to the non-relational data files; and

data driver means, coupled to the data engine means, for accessing the data in the non-relational data file specified by the subset of data access commands received from the data engine means according to the metadata.

10. The system of claim 9 wherein the dictionary means includes

field means for storing field definitions for each of the non-relational data files, the field definitions specifying the basic units of data as fields in the corresponding non-relational data file.

11. The system of claim 10 wherein the dictionary means includes

relation means for storing relation definitions for each of the non-relational data files, the relation definitions specifying compositions of fields as relations in the corresponding non-relational data file.

12. The system of claim 11 wherein the dictionary means includes

view means for storing view definitions for each of the non-relational data files, the view definitions specifying compositions of relations as views in the corresponding non-relational data file.

13. The system of claim 9 wherein the dictionary means includes

a dictionary containing metadata for at least one of the non-relational data files, and

a dictionary driver associated with the dictionary and used for accessing the metadata in the dictionary.

14. The system of claim 13 wherein the dictionary driver includes

a dictionary function vector defining metadata access functions supported by the dictionary driver, and wherein the data engine means includes

dictionary function vector means for communicating with the dictionary driver to obtain information regarding the metadata for decoding the queries into the subset of data access commands.

15. The system of claim 9 wherein the data engine means includes

parser means for ensuring the correctness of the relational-type queries and for forming Boolean expressions from the relational-type queries.

16. The system of claim 9 wherein the data engine means includes

optimizing means for placing the subset of commands into a sequence which expedites the access of the desired data.

17. The system of claim 16 wherein the optimizing means includes

means for organizing the Boolean expressions into a tree structure.

18. The system of claim 16 wherein the optimizing means includes

means for traversing the tree structure to form an access plan containing the sequence of commands to access the desired data.

19. The system of claim 18 wherein the optimizing means includes

means for examining the amount of system resources used by different sequences of commands.

20. The system of claim 16 wherein the data engine means includes

request processing means for executing the commands in the access plan.

21. The system of claim 20 wherein data driver means includes

a data driver function vector defining data access functions supported by the data driver means, and wherein the request processing means includes

function vector means for communicating with the data driver means to obtain information regarding the supported data access functions.

22. The system of claim 21 wherein the request processing means includes means, coupled to the function vector means, for executing the subset of data access commands in the access plan based on the information obtained regarding the metadata for decoding the queries into the supported data access commands.

23. The system of claim 21 wherein the function vector means includes

means for receiving from the data driver means a function vector identifying each of the data access functions supported by the data driver means and a corresponding address for invoking each of the supported data access functions, and wherein the request processing means includes

means for causing the data driver means to invoke the functions supported by the data driver means by transferring control to the corresponding addresses in the function vector.

24. The system of claim 9 wherein the data driver means includes

means for sending a function vector to the data engine means, the function vector identifying the data access functions supported by the data driver means and a corresponding address for invoking the supported functions.

25. The system of claim 9 wherein the non-relational data files have file types, and wherein the data driver means includes

a plurality of drivers each corresponding to a different non-relational file type, and wherein the data engine means is coupled to each of the drivers.