Method of integrating schemas of distributed heterogeneous databases

• US 5,806,066 A
• Filed: 03/26/1996
• Issued: 09/08/1998
• Est. Priority Date: 03/26/1996
• Status: Expired due to Term

• Current Assignee: Bull HN Information Systems Incorporated (Atos SE)
• Original Assignee: Bull HN Information Systems Incorporated (Atos SE)
• Inventors: Golshani, Forouzan, Howell, Thomas H., Friesen, Oris D.
• Primary Examiner(s): Lintz, Paul R.
• Assistant Examiner(s): Min, Donald
• Application Number:
  US08/624,726
• Time in Patent Office: 896 Days
• Field of Search: 395/601, 395/610, 395/611, 395/613, 395/614, 707/100, 707/102, 707/103
• US Class Current: 1/1
• CPC Class Codes: G06F 16/211 Schema design and management
• CPC Class Codes: G06F 16/289 Object oriented databases
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method for integrating the schemas of a plurality of independent and heterogeneous database management systems of a distributed database management system (DDBMS). The DDBMS includes a computer system in which the DDBMS resides and one or more subservient computer systems. The schemas of two of the independent database systems are fetched from the subservient computer systems. The schemas are converted from a relational database form to an object-oriented form. The schemas are then normalized and displayed graphically. Equivalencies are identified and the two schemas are integrated. These steps are repeated until the schemas of all data bases to be integrated have been integrated into a single integrated, or global schema. The global schema is then converted from the object-oriented form to the relational form, and SQL commands are created to allow data from subservient databases to be obtained to create a virtual database residing in the host computer system satisfying the requirements of the global integrated schema.

Claims

1. In a distributed database system (DDBS) having at least two independent database systems (DBS)s with each DBS including in addition to its database, a database management system (DBMS), and a system catalogue which includes a schema of the database, a representation of the structure of the database defining the kinds of data stored in the database but excluding the actual data;

the DDBS including a distributed data access (DDA) software module which provides a communication capability between the DDBS and the independent DBSs;

the method of producing a global schema including desired objects and associated attributes from the schemas of the two independent DBSs, the improvements comprising the following steps;

a, obtaining the schema of each of said independent DBSs;

b, identifying primary keys, foreign keys and data dependencies of the schemas of each of the two independent DBSs;

c, converting the schemas of each of the two independent DBSs to schemas in object-oriented form;

d, normalizing each of the object-oriented forms of the schemas produced in step c;

e, defining equivalence of objects of the two normalized schemas produced in step d;

f, integrating the two object-oriented schemas to produce a global object-oriented schema;

g, converting the global object-oriented schema of step f to produce a global schema in a relational form;

h, developing SQL commands for realizing the global schema within the DDA as a virtual database system satisfying all the requirements outlined in step g for accessing data from the independent database systems.

2. The method of claim 1 in which the independent database systems are in relational form.

3. The method of claim 2 in which each of the two independent database systems are heterogeneous.

4. The method of claim 3 in which step d further includes graphically displaying the normalized object-oriented forms of the schemas.

5. The method of claim 4, in which step g, further includes eliminating extraneous data from the global object-oriented schema produced in step f.

6. The method of claim 5 in which the distributed database system resides in a computer system including a memory system.

7. The method of claim 6 in which the global relational database schema produced in step g and the SQL commands developed in step h are stored in the memory system of the host computer system.

8. In a distributed database system (DDBS) having at least two independent heterogeneous database systems (DBS)s with each DBS including in addition to its database in a relational database form, a database management system (DBMS), and a system catalogue which includes a schema, a representation of the structure of the database defining the kinds of data stored in the database but excluding the actual data;

the DDBS includes a distributed data access (DDA) software module which provides a communication capability between the DDBS and the independent DBSs;

the method of producing a global schema including desired objects and associated attributes of the two independent DBSs, the improvements comprising the following steps;

a, obtaining the schema of each of said independent DBSs;

b, identifying primary keys, foreign keys and data dependencies of the schemas of each of the independent DBSs;

c, converting each of the schemas from a relational database form to an object-oriented form of a database;

d, normalizing each of the object-oriented forms of the schemas produced in step c;

e, producing a graphical display of the normalized object-oriented forms of the schemas produced in step d;

f, defining equivalence of objects of the two schemas displayed in step e;

g, integrating the two object-oriented schemas to produce an integrated object-oriented schema;

h, converting the integrated object-oriented schema produced in step g to an integrated schema in relational database form;

i, eliminating extraneous data fields from the integrated schema produced in step h; and

j, developing SQL commands for obtaining data from the databases of the independent database systems realizing the global schema within the DDA as a virtual database system satisfying the requirements outline in step h for accessing data from the independent database systems.

9. The method of claim 8 in which the distributed database system includes a computer system in which the distributed database system resides, said computer system including a memory system, and subservient computer systems in which the independent heterogeneous database systems reside.

10. The method of claim 9 in which the integrated schema at the completion of step i and SQL commands developed in step j are stored in the memory system of the computer system in which the distributed data base system resides, and from which they are retrieved as needed.

11. A method by which a distributed database system (DDBS) which includes a computer system in which the DDBS resides, and at least "n" independent database management systems (DBMS)s with each independent DBMS residing in a computer system, where "n" is an integer greater than 1;

each of the "n" independent DBMSs including a database (DB), and a system catalog which includes a schema of the independent DBMS database, where a schema of a given DB is a representation of the structure of the DB defining the data stored therein but excluding actual data;

the computer system in which the DDBS resides including a distributed data access (DDA) software module for providing communications between the computer system in which the ddbs resides, the computer systems in which the independent DBMSs reside, and the independent DBMSs residing therein for producing a single integrated schema from the "n" independent database management systems;

comprising the steps of;

a, after selecting a computer system in which a first independent DBMS resides, obtaining the schema of that independent DBMS and displaying said schema;

b, after identifying primary keys, foreign keys and data dependencies of the displayed schema, transforming the schema to a "Third Normal Form" if not in that form as displayed;

c, transforming the schema in Third Normal Form to an object-oriented DB form;

d, displaying the object-oriented DB schema;

e, repeating steps a-d with respect to a second independent DBMS;

f, after specifying correspondences between the two displayed object-oriented DB schemas and eliminating conflicting assertions, merging the two object-oriented schemas to produce an integrated schema in objected-oriented form;

g, repeating steps a-d, and f on the remaining independent DBMSs until the schemas of all independent DBMS, have been merged to form a single integrated schema;

h, transforming the integrated object-oriented DB schema at the completion of step g to an integrated schema in relational form;

i, generating Database Language SQL commands to create a view of the integrated schema in relational form; and

j, storing the merged integrated schema of step i and the SQL commands generated in step i in a memory system of the computer system in which the DDBS resides.

12. The method of claim 11 in which the independent database management systems are relational database systems and are heterogeneous.