Join Martin Guidry for an in-depth discussion in this video Introduction to business intelligence, part of Implementing a Data Warehouse with Microsoft SQL Server 2012.
In this section I'd like to talk about Business Intelligence and how it relates to data warehouses. Concept of Business Intelligence is actually a fairly old idea going back about 150 years. It was originally defined as gaining a competitive advantage by your ability to gather information more quickly than your competitors. The modern definition still focuses on gathering information. Quickly is still important, but many business seem to be focusing more on a large quantity of data, rather than getting it quickly.
In the modern marketplace, the definition of business intelligence varies a little from one vendor to another vendor. Each of them are trying to frame the problem in a certain way that they're product solves that problem. There are common characteristics amongst almost all definitions of business intelligence. We're typically looking to find meaningful information in raw data. We want to use that data to support operational decisions about the business. We want to use the data to measure the health and success of the business. And in some cases we try to perform simulations on the data to predict the future.
So all of these goals involve data, involve doing something to data. Where is that data going to come from? Well, we have a few different choices. In an application designed to analyze or visualize the stock market, you would probably get real time data streaming into the computer, not stored in the database. We could some business intelligence by searching individual documents. That's pretty rare and very time consuming. But it is plausible that our business intelligence process could search individual Word or PDF documents.
Some light weight and prototype Business Intelligence implementations involve using transactional data bases that are also being used for other things. This can cause some problems typically a transactional database will not perform very well when a large number of reports are being run against it. And adding the additional overhead of a BI infrastructure to an existing database can be very taxing on that database. So the best solution for data storage for most BI scenarios is a data warehouse.
And I would say, in my experience, over 99% of enterprise level BI solutions do involve a data warehouse. The data warehouse is sometimes the only data source with some of these other options being used as smaller secondary data sources. Now just having the Data Warehouse doesn't give us a for BI solution. We're still going to need other pieces. The two parameter pieces Microsoft provides are a reporting tool known as Microsoft Sequel Server Reporting Services.
And an analysis tool. Known as Microsoft SQL Server Analysis Services. We're going to talk about both of these in the upcoming sections.
- Typical databases vs. vdata warehouses
- Choosing between star and snowflake design schemas
- Working with appliances and "big data"
- Exploring source data
- Implementing data flow
- Debugging an SSIS package
- Extracting and loading modified data
- Enforcing data quality
- Consuming data in a data warehouse