Data Warehousing

The terms “data warehouse” and “data warehousing” can be used synonymously. They both refer to a database that pulls data from multiple, more specialized databases being used within an enterprise environment. There are multiple ways for a data warehouse to be deployed, depending on the IT environment and needs of an organization.

The larger an organization gets, the more unruly the data collection can be. A data warehouse can be developed to provide several advantages for a mid- to large enterprise. The key advantages are speed of data processing, reducing errors and redundancy database tables, and access to large data sets for analysis.

For small organizations, a database may contain all of the different functions they need to conduct business. Accounting, customer service, and inventory controls could all be in the same database because there is not a lot of data generated or a consistent use of the database. The database can share tables more effectively and information can be more general. As an organization increases in size, the use of the database increases also, causing a latency problem. Creating a data warehouse allows an organization to separate databases from the whole of the organization to reduce the latency within a system.

As an organization grows, different functions within the organization become more specialized. Different functions (departments) develop more specialized databases that provide greater control over their activities. Specialized databases create more data as well. However, creating separate databases for the different departments creates an opportunity for redundancy and errors within the different databases. Data warehousing provides access to tables with common information that may be shared throughout the different functions within an organization, reducing latency, and reduces the possibility for errors and redundancy.

Analyzing and generating reports is easier with data warehousing as well. A data warehouse pulls data from all of the specialized databases that are being used by different departments. An analyst can search for, and develop reports on, various relationships between departments and the organization itself.

For larger enterprises, there may be a “data mart” as well. A data mart is a type of data warehouse that collects and manages data based on organizational function. For example, an accounting department may be divided into different segments, such as payables and receivables. Each may have their own database for their specialized function. The accounting department would have a data mart that pulls information from payables’ and receivables’ databases for accounting management to analyze. The data warehouse would pull data from the accounting data mart for C-suite management to analyze.