Data architecture for fetch-intensive database applications

• US 6,006,216 A
• Filed: 07/29/1997
• Issued: 12/21/1999
• Est. Priority Date: 07/29/1997
• Status: Expired due to Term

• Current Assignee: WSOU Investments, LLC (WSOU Holdings, LLC)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Hull, Richard Baxter, Griffin, Timothy G., Zhou, Gang, Lieuwen, Daniel Francis, Kumar, Bharat
• Primary Examiner(s): Amsbury, Wayne
• Assistant Examiner(s): Loomis, John C.
• Application Number:
  US08/902,447
• Time in Patent Office: 875 Days
• Field of Search: 707/2, 707/4, 707/101, 707/104
• US Class Current: 1/1
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A logical data architecture for fetch-intensive database applications includes a first and a second logical database. The second database includes materialized views over the first database. Updates are propagated from the first database to the second database using so-called view maintenance techniques. In preferred embodiments, the first database may be optimized for update transactions and the second database may be optimized for read transactions. At least one of the materialized views included in the second database is based on a pre-defined application read transaction. The update processing overhead associated with view maintenance is substantially performed at the first database.

Claims

1. A method of operation of an on-line fetch processing application, said method comprising

directing a majority of updates to a first database,

directing a majority of reads to a second database, said first and second databases being such that said majority of updates are performed more efficiently to said first database than they would be if said updates were to be performed to said second database, and said majority of reads are performed more efficiently from said second database than they would be if said reads were to be performed from said first database, said second database holding materialized views, said materialized views being based on said first database, and

substantially maintaining said second database by at least one technique of view maintenance.

2. The method of claim 1 wherein at least one of said materialized views is based on at least one application-specific read.

3. The method of claim 1 wherein at least one of said techniques of view maintenance is a technique of incremental view maintenance.

4. The method of claim 1 wherein update processing associated with occurrences of a data item in both said first and said second databases occurs substantially at the first database.

5. The method of claim 1 wherein said on-line fetch processing application being such that:

a) most transactions are read-only;

b) most of the read-only transactions use key-based selections; and

c) substantially all transactions are defined at system design time.

6. An apparatus comprising

a first database;

and

a second database including materialized views over said first database, said first and second databases being such that a majority of updates are performed more efficiently to said first database than they would be if said updates were to be performed to said second database, and a majority of reads are performed more efficiently from said second database than they would be if said reads were to be performed from said first database, at least one of said materialized views being based on at least one read specific to an application, said application being an on-line fetch processing application such that;

a) most transactions are read-only;

b) most of the read-only transactions use key-based selections and

c) substantially all transactions are defined at system design time;

wherein a majority of updates to said second database are performed via at least one technique of view maintenance.

7. The apparatus of claim 6 wherein processing associated with said view maintenance is substantially performed at said first database.

8. The apparatus of claim 6 wherein said apparatus is a telecommunications switch.

9. The apparatus of claim 6 wherein said technique of view maintenance is a technique of incremental view maintenance.

10. The apparatus of claim 9 wherein said technique of incremental view maintenance includes communication of at least one delta to said second database.

11. The apparatus of claim 6 wherein either or both of said first and said second databases are physically distributed across platforms.

12. The apparatus of claim 6 wherein both of said first and said second databases reside on the same platform.

13. The apparatus of claim 6 wherein said first database is more normalized than said second database.

14. The apparatus of claim 6 wherein said first database is optimized for at least one application-specific update.

15. An apparatus comprising

a first database structured so as to optimize at least one update specific to an application, said application being an on-line fetch processing application;

and

a second database having materialized views, said views being based on said first database, said second database being defined so as to optimize at least one read specific to said application, said second database being maintained by at least one technique of incremental view maintenance.

16. The apparatus of claim 15 wherein said first database includes means for computing updates to said materialized views and for applying them to said second database.

17. The apparatus of claim 15 wherein the first and second databases are such that a majority of updates is performed more efficiently to said first database than they would be if said updates were to be performed to said second database, and a majority of reads is performed more efficiently from said second database than they would be if said reads were to be performed from said first database.

18. The apparatus of claim 15 wherein said o n-line fetch processing application being such that:

a) most transactions are read-only;

b) most of the read-only transactions use key-based selections; and

c) substantially all transactions are defined at system design time.

19. A telecommunications switch comprising

a first database;

and

a second database including materialized views over said first database,

wherein a majority of updates to said second database are performed via at least one technique of view maintenance.

20. The switch of claim 19 wherein said first database includes means for computing updates to said materialized views and for applying them to said second database.

21. The switch of claim 19 wherein said second database is defined using at least one application-specific query over said first database.

22. The invention of claim 19 wherein said first and second databases being such that a majority of updates are performed more efficiently to said first database than they would be if said updates were to be performed to said second database, and a majority of reads are performed more efficiently from said second database than they would be if said reads were to be performed from said first database.

23. The invention of claim 19 wherein said technique of view maintenance includes incremental view maintenance.

24. A telecommunications switch comprising

a first database;

and

a second database including materialized views over said first database, the first and second databases being such that a majority of updates are performed more efficiently to said first database than they would be if said updates were to be performed to said second database, and a majority of reads are performed more efficiently from said second database than they would be if said reads were to be performed from said first database, at least one of said materialized views being based on at least one application-specific read;

wherein propagation of said majority of updates to said second database is performed via at least one technique of incremental view maintenance, the processing associated with said view maintenance being substantially performed at said first database.