Do you need a Data Warehouse?
Reports are everywhere and self-service reporting has been popular for years. But where is the source data coming from? Is it from queries directly against your line of business applications? If so you may want to consider implementing a data warehouse. The tools and technologies available today make that choice much easier to make, plus much more affordable. And you'll gain significant benefits from having your business data properly staged and managed.
Not so many years ago data warehouses were expensive repositories of certain corporate data that involved long, expensive projects and big funding.
That has changed in the last few years with some of the new business intelligence technologies released by Microsoft and other vendors. Now they are much more affordable and can be created much more quickly.
That still leaves the 3 big questions:
-
do you need a data warehouse?
-
what should it 'look like'?
-
should you create it yourself?
As with most technology decisions it requires consulting and analysis to make the decision.
As part of Microsoft self service BI efforts we now have more choices in data warehousing. SQL Server 2012 brought a new type of analytical processing model known as the Tabular Model. This model is much more accessible to companies and the learning curve is less steep than the traditional Multidimensional Analysis Server.
The tabular data model is based on tables and relationships, just like a traditional OLTP system. However, they are presented to client reporting applications as cubes with measures, dimensions, calculated fields, etc.
Many of Microsoft's client tools, such as Excel and Power BI, can use the tabular data model as a data source. Excel PowerPivot can even create the model!