What’s Wrong With My Data? - High-Performance Web Databases: Design, Development, and Deployment - page 414

Redundant Data with Inconsistencies

In Exhibit 27-2:

• Our hypothetical data warehouse draws information from both Files
“A” and “B” (assume each has unique fields of interest not displayed in
the above example). Because each file was developed and used sepa-
rately by different divisions of the company, redundant information is
stored. Which will you use as your primary source for those fields
available from both?
• Note that the information for Jones is not consistent. Is it safe to as-
sume that File “A” has more recent information, and therefore that
Jones took up smoking at age 23? Regardless of the answer, how do we
account for the difference in

for Lee?
• Data File “B” has a record 525 not present in File “A.” Should we then
merge the records from the two files to get a complete set? Or perhaps
File “B” is out of date, and the record should have been deleted from
File “B” as well? This is another illustration of inconsistency.

Sex

Exhibit 27-2. Example of redundant data.

Data File “A”

Identifier Last Name Sex Age Smoker

523 Smith M 00 N

524 Jones F 23 Y

526 Lee M 42

527 Frank 17 Y

528 Yu M 00 N

Data File “B”

Identifier Last Name Sex Age Smoke

523 Smith M 00 N

524 Jones F 22 N

525 Samuelson M 54 Y

526 Lee F 42

527 Frank 17 Y

Different/Changing Business Rules or Logic

In Exhibit 27-3, our hypothetical data warehouse draws information
from both Files “A” and “C”:

• Notice that Lee's status is “Y” in File “C.” This is a result of a

business rule in the application logic which says “Unless the individu-
al specifically indicates that they are a non-smoker, give them the

Smoker