What is the difference between 1.0, 2.0 and 3.0 and why are the differences important. What follows is an illustration that may provide some understanding.
Data Warehouses 1.0 (the Data Mart) are often referred to as a “subset” of a data warehouse; unfortunately data marts, and, in fact, instances of data warehouses 1.0, are no more than a SQL table representation of the processing system that is providing the data, such as a trading system, or accounting system. These data marts were born from the generally “poor reporting” of the originating processing engines created in the last century. The benefit of a data mart is that you can report, but you are limited to silos of data with extremely limited navigation (i.e. you couldn’t navigate from performance data to position data, to tax lots, or to allocations, as the data would be in 3 different data marts with different references, and different keys and foreign keys. Put simply, when creating a data mart, a developer typically takes an existing extract from a processing engine (e.g. trade history) and loads it into a SQL table verbatim, with minimal or no “warehouse” functionality.
Find out more about Data Warehouses 2.0 and 3.0: http://investcloudsolutions.blogspot.com/2012/05/data-warehousing-30-understand.html