home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!portal!cup.portal.com!DVaz
- From: DVaz@cup.portal.com (David D Vazquez)
- Newsgroups: comp.databases
- Subject: Re: Normalizing 2NF -> 3NF
- Message-ID: <69674@cup.portal.com>
- Date: Mon, 16 Nov 92 22:02:05 PST
- Organization: The Portal System (TM)
- References: <1992Nov15.163413.17527@news.uiowa.edu>
- <1992Nov15.223825.16263@odin.diku.dk>
- <1992Nov16.103334.15543@neptune.inf.ethz.ch>
- <1992Nov16.135135.3108@odin.diku.dk>
- Lines: 32
-
- marti@nugget.inf.ethz.ch (Robert Marti) writes:
- . :
- . FDs: A -> B and B -> C
- . Relation: R(A*,B,C)
- . Problem: How to decompose?
- . Suggestion: R1(A*,B) and R2(B*,C) or R1(A*,B) and R2(A*,c)
- . ^^^^^^^^^^^^^^^^^^^^^
- .Yes, but this might not be efficient. Fx, if there are two groups of users
- .that wants to see an (A,B) and an (A,C) view resp., it requires a lot of
- .unnecessary joins. Eg, (A,B,C) == (Name,Address,Phone#). You typically look up
- .addresses from names and phone-numbers from names---not phone# from addresses.
- .
-
- It may make that specific query faster, but it can wind up costing you more
- disk space. While R1(A*,B) and R2(A*,C) are in 3NF given the above stated FDs,
- you will have alot of redundant (repeated) data (3NF is supposed to solve this
- correct?).
-
- .A DBMS should be able to check other integrity constraints than primary keys.
-
- Correct, but a 'database design' should enforce all of the given functional
- dependencies (R1(A*,B) + R2(A*,C) will not).
-
- .--
- . Niels Elgaard Larsen
- . Institute of Computer Science,
- . University of Copenhagen
- . E-mail: elgaard@diku.dk
-
- David Vazquez
- Computer Science major @ University of New Orleans
- E-mail: dvaz@cup.portal.com and dvazquez@sdiserv.cs.uno.edu
-