Data warehouses play a pivotal role in business intelligence (BI). A data warehouse is a central repository of business data. It copies information from various business databases, consolidates it, and stores it in a format that provides high performance querying and analysis. The data warehouse thereby becomes a single source of data that serves consistent information. In an effort to keep business applications running smoothly, many organizations purge their business databases periodically; data warehouses also preserve this historical information.

As a business’s central data store, a data warehouse serves a variety of purposes and audiences. Monthly financial statements can pull from the same data warehouse that serves up the marketing department’s daily campaign performance updates and the sales organization’s quarterly revenue and quota reports. From this position, the data warehouse provides comprehensive analytical query capability to business decision makers, who can observe how all aspects of their company are performing as well as how the company is changing and growing over time.

Data warehouses provide analytical capability from the top of an organization to the bottom. In practical terms, this means that your high-level company information is supported by the detailed data that comprises it. An executive can analyze overall company sales, while a sales manager reviews a sales region and a branch manager views the sales from their office, each querying this information from the same data warehouse. Perhaps most importantly, the data warehouse enables data from different business functions, such as production and sales, to be used in the same report to provide a view of the company that is not only accurate, but complete and multifaceted. Data warehouses, working in concert with other business intelligence tools such as OLAP, make this information available to all of these users quickly and easily.

Technology Overview

Data warehousing is comprised of two layers of technology: the data extraction, transformation, and loading (ETL) layer, and the storage layer, which houses data in a SQL database.

The first component of a data warehouse, the ETL (Extract, Transform, & Load) layer, handles the extraction of data from various source databases, the transformation of that data into a structure specifically designed for analysis, and the load of that structured data into a repository. It is the transformation task that differentiates ETL from other forms of data migration. Transformation involves cleansing source data, such as excluding unwanted or “noise” data, and consolidating it with data from other systems by standardizing descriptive attributes (for instance, correcting product IDs that differ between databases). ETL also applies standardized calculations, such as currency conversions, unit conversions, and profit and business metric calculations to lines of data as they arrive from source systems.

For more detailed information about Data Warehousing and ETL services offered by Northridge Systems, please send us an information request or email sales@northridge.com