Table of Contents
Before You Create a Data Source
Send comments on this topic.
Before you create a Data Source, consider the following questions:

 

Do you need to place data from different sources/tables on the same data visualization widget (such as a chart or pivot table)?

A data visualization widget can be bound to a single Data Source. If you need to place data from multiple tables or files on a single widget (a pivot table or a chart), creating a separate Data Source for each will not do the trick.

 

If your data is placed in separate tables, but in the same database server you have two options:
  1. Create a view with the SQL query the merges the data as required and then create a single Data Source over this view.  When large amounts of data and/or lots of queries are required - this may not be a feasible approach.
  2. Create an ElastiCube that merges the data and create a single Data Source over this ElastiCube.

 

Do you need to query a lot of data (500K rows of data and higher)?

Some databases perform better than others when asked to service queries over large data.   This becomes even more apparent when multiple users are querying the data simultaneously.  If the database you are using cannot handle the query requests, you should considering importing the data into an ElastiCube and creating the Data Source over it.  ElastiCubes are specifically designed for large data volumes and numerous requests.