It ' s import to know before start learning Dw/bi

Data warehousing and business intelligencedifferences between data warehousing and business intelligencefiled Under:busin ESS intelligence,data warehousing-vincent Rainardi @ 6:38 pm
Tags:business Intelligence, Data warehousing

Try asking your colleague what's the difference between business intelligence and a data warehouse. I find that a lot of people, even those who work in BI projects and bi industry, does not understand the difference. A lot of people use these 2 terms interchangeably. Some people even prefer to use 1 term instead of the other because it simply "sounds better". Many people think that business intelligence was not just a data warehouse, but there was more to it. But when asked ' what business intelligence systems is not data warehouse systems? ' or ' What part of the business intelligence The most of them has difficulties explaining the answer. Warehouses.

These days, ' business intelligence ' is the norm used by most vendors in the industry, rather than "Data Warehouse". Most of the them call/classify their tools as business intelligence software, not data Warehouse software. The name of Cognos product is ' Cognos 8 business Intelligence '. BusinessObjects label themselves as "Bi software Company" and "global leader in BI software". The name of one of Hyperionproducts is "Hyperion System 9 bi+". SAS Enterprise BI Server provides a fully integrated and comprehensive suite of business intelligence software. Microsoft promotes SQL Server 2005 as the End-to-end Business intelligence platform. It seems that is Kimball Group who consistently with the term data warehouse. Bill Inmon, as the inventor of this term, also uses the term Data warehouse.

So, let's get into the details. This is an example of a data warehouse system:

It includes ETL from the source system, front end applications (those boxes on the right hand side), and everything in Between. It has a control system, an audit system and a data quality system (also known as data firewall). Not all data warehouse systems has all of the components pictured above, for example, some data warehouse system could not hav e Operational data stored (ODS), see this article for details.

The 2 blue items are data Warehouse databases. The cylinder is in relational format (labelled as dimensional data store, DDS for short), the box was in multidimensional F Ormat (labelled as cubes in the picture above). This blue cube was also known as on Line Analytical Processing cube, or OLAP cube for short.

The yellow items is business intelligence applications. Most business intelligence applications take the data from multidimensional format Data Warehouse, but some does take data from The relational format. The whole diagram above is also known as Business Intelligence System.

Some Business intelligence applications take data directly from the source system. For example, some dashboard systems could get sales summary data from the source system and display it in gauge meter format . In this case, we can not call the system a data warehouse system. It's still a business intelligence system, but it's not a data warehouse system, because it does not has a data Warehou SE database behind the gauge meter application.

Business intelligence systems, in the past also known as executive information systems, or decision support systems, AR e a non-transactional IT system used to support business decision making and solve management problems, normally used by T OP executives and Managers. many varied definitions exist in the market place today about the business Intellige nCE system; one from Dr Jay Liebowitz is arguably one of the better ones. Most people agree, OLAP and Data Warehouse systems is a major and important part of the business intelligence systems. Most business intelligence systems is in the form of a data warehouse systems. Yes, there is business intelligence systems, does not use OLAP or data warehouses, as illustrated in the example of Gau GE meter application above, but they is more rare than the ones with OLAP or a data warehouse.

According to Ralph Kimball, in the He book the data Warehouse ETL Toolkit, a data Warehouse are a system that extracts, cleans , conforms, and delivers source data into a dimensional data store and then supports and implements querying and analysis For the purpose of decision making. He stressed that a data warehouse are not a product, a language, a project, a data model or a copy of the transaction system. In a interview with Professional Association for SQL Server (PASS) on 30th April 2004, he explained about the Relationshi P between data warehousing and business intelligence.

In their latest book, The Microsoft data Warehouse Toolkit, Joy Mundy and Warren Thornthwaite does not differentiate Data WA Rehouse systems and business intelligence systems. They consistently use the term DW/BI system throughout the book. This was understandable because, as I describe above, most business intelligence systems was in the form of a data warehous E system.

Bill Inmon, who invented the term data Warehouse, defines data warehouse as a source of data that's subject oriented, Integrated, nonvolatile and time variant for the purpose of management ' s decision processes. He pointed the term Data warehouse is never trademarked or copyrighted. As a result, anyone can call anything a data warehouse. He recently defined a new term, DW 2.0, and this one are trademarked so nobody can change the definition. He explained the Architecture in his article in Dmreview, along with the differences between the first generation of Data warehouses and DW 2.0 and its advantages.

So, as a summary, back to the original question, what's the difference between data Warehouse and business Intelligenc E? Most business intelligence systems is based on data Warehouse systems (the one with dimensional model, fact tables, dimen Sion, etc), but some business intelligence systems is not data warehousing, i.e. taking data directly from the source Sys TEM, like the example described above. Business Intelligence application (as opposed to business intelligence system) are the yellow boxes on the diagram above, I . e. The front end applications. The Data Warehouse database (or sometimes people dropped the word database, so it becomes just ' data Warehouse ') is the BL UE Cylinder and blue box on the diagram above, i.e. the dimensional storage, whether in relational database format or in M Ultidimensional database format.

If people say ' Data warehouse ', be careful because it can mean either Data Warehouse system (the whole diagram above) or D ATA Warehouse Database (just the blue items). If people say ' business intelligence ', it can mean either business intelligence system (the whole diagram above, or a BI s Ystem without Data Warehouse) or Business intelligence application (the yellow boxes).

I hope this article makes the terms clearer, but I am open to comments and suggestions. As Ralph Kimball said, if you ask the different people what the data warehouse is that you were likely to get the different answers.

Vincent Rainardi
1st May 2006

This was a repost from sqlservercentral.

It ' s import to know before start learning Dw/bi