Considering the business requirements of the Data Warehouse
An Introduction to Data Warehouses and Data Warehousing Part 2 - Considering the business requirements of the Data Warehouse
Before you can start gathering your data in the data warehouse you should undergo a thorough analysis of your data warehouse requirements. Essentially this means you should be asking questions such as:
- What summarised information do I want to be able the data warehouse to produce?
- What departments will be using the Data Warehouse
- How far back do I want to be able to analyse data i.e. last 3 years or less?
- What will happen to data once it is no longer required
- When will updates to the data take place? How much downtime (if any) can be afforded for updating data / maintenance?
- Is query speed more important than storage costs?
Now go back and look at the first point in the list again and start to talk to the people who will be using the front end analytical tools. Do they know what information they want to be able to view. Do you already record this data? You can't create summaries if you don't have the raw data that is required to create the summary.
This identification of business requirements and expected report specifications is key in the success of the Data Warehouse. It cannot be emphasized enough how important this stage is. If you get it wrong, after spending a lot of time and effort building the data warehouse you will find that it is not fit for purpose and the end users will be very unhappy.
The maintenance of a Data Warehouse is an ongoing process and new data will have to be imported at set time intervals, be it weekly, monthly, quarterly etc.
The Data Life Cycle and archiving
In your analysis of needs you will identify the Data Life Cycle that occurs in various areas of your companies data use. The archiving or deletion of old data is something that must be considered when designing your data warehouse and even in these days of cheap storage media, the capacity of your storage should be considered.
In the next part of this series on an introduction to Data Warehouses and Data Warehousing we will look at designing the Data Warehouse and the concept of dimensional data modelling.