Database Reference
In-Depth Information
CUSTOMER MASTER FILE
CustNo
CustName
Address
Country
1000
Allbooks Book Store
5757 Westheimer, Houston, TX 77057
U.S.A.
1010
Akito Books
Chiyoda-ku, Tokyo 100
Japan
1040
Robert Smith Ltd.
10 Bonds St., London W1A 2AA
U.K.
2050
Sally Orobetz
8 Hazelton Ave., Toronto, Ontario M5R 2E1
Canada
BOOK MASTER FILE
ISBN
PUBLISHER FILE
PubId
Title
Author
PubId
PublisherName
Country
2093356790
2101155897
1558712215
3456765432
DB Design
Carey
100
200
300
400
100
200
300
400
Ron Fairchild
USA
DW Fundamentals
McMillan
Crosley
U.K.
Art Appreciation
Stewart
Summer Hill
Canada
Existentialism
Ernst
Ully Wille
Germany
SALESPERSON FILE
SalRepId
SalRepName
Office
Comm%
10
20
30
40
Williams
Chicago
13
12
9
10
Harreld
London
Swamy
Toronto
Katzman
Munich
SALE TRANSACTION FILE
SalDate
PAYMENT TRANSACTION FILE
PymtDte
CustNo
PayMethod
Amount
CustNo
SalRepId
ISBN
Qty
Amount
13-Apr
1000
1010
1040
2050
Amex
600.00
6-Apr
1000
1010
1040
2050
10
20
30
40
2093356790
2101155897
1558712215
3456765432
10
10
20
10
799.50
22-Apr
Visa
500.00
10-Apr
699.50
24-Apr
Check
800.00
19-Apr
1,000.50
30-Apr
Visa
650.00
24-Apr
750.00
Figure 1-4
Progressive Book Distributors: files.
the end of a month. All the sales transactions and payment transactions for the
month had to be batched together and processed in a batch mode.
Figure 1-5 indicates a flowchart of the jobs that must be run to produce the billing
statements.
Even an initial review of the flowchart reveals that there are too many sorts for
just producing simple billing statements. That is because each file is kept in a
sequence that is not useful for the processing logic of the entire application. The
whole concept of batch processing is very inflexible. Suppose in the middle of the
month you need the billing statements for just a few select customers. There is no
easy way of doing this. You have to run the complete batch process in the middle
of the month and then separate out the statements you need.
File-oriented systems are inadequate to face the challenges of increasing demand
for information. Especially when companies care for information as a key asset, the
earlier file-oriented data systems possess severe limitations. Let us discuss the
important limitations so that we can appreciate how database systems overcome
these shortcomings.
Uncontrolled Data Redundancy In file-oriented systems, each application has
its own set of files. Each application creates and stores data just for the use of that
application. For example, in a bank, there could be three separate applications—
one for checking accounts, one for savings accounts, and another one for loan
Search WWH ::




Custom Search