Data Warehouse or Business Intelligence BI Solution – or Both?

I’ve recently been working with a number of small- and medium-sized businesses that are preparing to jump aboard the decision support bandwagon. Several of them have been asking me questions that reveal the enormous amount of confusion in the decision support industry, such as, “do we need a data warehouse, or should we buy a business intelligence solution?” That’s a question best answered by clearly defining some of the key terms that the industry throws around.

Business Intelligence, or BI, is another name for decision support. In essence, BI refers to the business-level practice of gathering information in such a way that management and decision-makers within the organization can quickly get the answers to key business questions. In other words, BI is a generic name for an entire set of functionality as well as the supporting technologies.

A data warehouse is one possible component of a BI system. The idea is that most companies store data in many different places: Customer databases, order databases, shipping databases, and more. These systems can be slow to query individually, and answering key business questions may require data from several sources to be combined in unique ways. A data warehouse basically copies all of that data into one place, making it easier and faster to query and combine that data.

In-memory analytics achieves more or less the same thing as a data warehouse, but it doesn’t require as much advanced planned. Whereas the design of a data warehouse requires you to think in advance about how you want to combine data, in-memory analytics can retrieve and combine data on the fly. As the name suggests, these analytics take place in the memory of a server, meaning the server typically needs relatively massive amounts of memory and processing power. The idea of data warehouses were created before this amount of memory and processing power were readily available.

A BI solution is a software package that implements a data warehouse, in-memory analytics, or both, along with a front-end system that allows users to interact with the system. That front-end usually incorporates dashboards, reports, scorecards, and other end-user elements.

You’re never going to have just a data warehouse, although you may well have a BI solution that only utilizes a data warehouse. These days, it’s more likely that you’ll have a BI solution that leverages both a data warehouse and in-memory analytics, since the two technologies are complementary. The data warehouse can make it easy to quickly get standardized answers to predictable questions, while in-memory analytics serves as kind of an “ad-hoc backup,” providing answers to questions that couldn’t be foreseen when the data warehouse was constructed. An in-memory analytics system can also draw data from the data warehouse, as well as other sources.

The real moral here is to not focus at all on the underlying technologies. If you’re considering a BI solution, look at its business-level features first, and don’t worry so much about how the solution delivers those features. Whether it uses a data warehouse, in-memory analytics, or pure magic under the hood, so long as it’s able to provide the capabilities your business needs, that’s all you should care about.

Don Jones
PowerShell and SQL Instructor – Interface Technical Training
Phoenix, AZ

Posted in SQL Server | Posted in , , , , | Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code class="" title="" data-url=""> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> <pre class="" title="" data-url=""> <span class="" title="" data-url="">