Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»

25.    Do the data structures internal to the data warehouse environment make use of the following:

■    Arrays of data?

■■ Selective redundancy of data?

■    Merging of tables of data?

■    Creation of commonly used units of derived data?

ISSUE: Even though operational performance is not an issue in the data warehouse environment, performance is nevertheless an issue. The designer needs to consider the design techniques listed previously when they can reduce the total amount of I/O consumed. The techniques listed previously are classical physical denormalization techniques. Because data is not updated in the data warehouse environment, there are very few restrictions on what can and can’t be done.


The factors that determine when one or the other design technique can be used include the following:

■■ The predictability of occurrences of data ■■ The predictability of the pattern of access of data ■■ The need to gather artifacts of data

26.    How long will a recovery take? Is computer operations prepared to execute a full data warehouse database recovery? A partial recovery? Will operations periodically practice recovery so that it will be prepared in the event of a need for recovery? What level of preparedness is exhibited by the following:

■    Systems support?

■    Applications programming?

■■ The DBA?

■■ The DA?

For each type of problem that can arise, is it clear whose responsibility the problem is?

ISSUE: As in operational systems, the designer must be prepared for the outages that occur during recovery. The frequency of recovery, the length of time required to bring the system back up, and the domino effect that can occur during an outage must all be considered.

Скачать в pdf «Building the Data Warehouse»