Creating a Filtered OLAP Data Cube

Introduction

One of the unique features of Argos is the ability to create filtered OLAP Data Cubes.  That is, information from a data source can be filtered before providing the information to the OLAP cube for analysis.  Also, OLAP cubes can be built off of ANY data source – you are not dependent on a data warehouse.  The underlying theory for OLAP Cubes is discussed in the Argos Report Viewer Guide,  so it will not be discussed here.

This example will create an OLAP Data Cube consisting of sales information contained in the sample database.  Items of interest include sales totals by region, by employee, by date, and by product type for all sales transactions.

Data will be filtered by a date (specified by the Report Viewer) before it is sent to the cube for analysis.  Even though date ranges could be filtered within the cube, for the sake of example, data will also be filtered by a date provided on the input selection form.  Only sales after the date entered on the form will be made available to the OLAP cube.

Tip: There are numerous sources that can be found on the web that discuss the underlying technology for OLAP.

Create the OLAP Object in the DataBlock Designer

1. Create a DataBlock in the Argos Explorer tree and select the appropriate connection. Edit the DataBlock which launches the Argos DataBlock Designer.   
2. In the DataBlock Designer, add a date field which will be used to filter the data.  The date will be used to exclude sales records prior to the date entered on the form.
   
3. Click the OLAP cube object in the toolbar, then click anywhere in the Design Area.  The skeleton of the OLAP cube will appear as shown in the figure below.  Click on the corners of the cube to expand the size as desired.
   
4. Double-click inside the cube to configure the main properties of the cube.
   
5. In this simple example, the default selections will be used. Click the Next button to build the Fact Table.