BIRT Exchange Forum: BIRT Exchange Forum -> BIRT Rocks!

If you have used Excel, you probably have used and liked the convenience of Pivot Tables to automatically sort, count and aggregate data displayed in a compact row and column matrix. Pivot Tables, also known as Cross Tab(ulations), helps you to analyze data and recognize patterns in it. While BIRT has been supporting crosstabs for a long time, in version 11 Actuate has added support to interact with a crosstab table and automatically create charts using just a web browser. The tool that enables this interactivity is called BIRT Data Analyzer and it is well integrated with Interactive Viewer.

A crosstab report leverages a multi-dimensional data structure called Cube that is optimized for analysis. A cube organizes data into Measures, Dimensions & Attributes and can be stored in BIRT Document or BIRT Objects (.data file) (See my earlier blog on BIRT Objects). You can launch up a cube stored in either format for analyses in Data Analyzer, take a look at the short video recording to see it in action:


See the above video in full screen.

Data Analyzer helps you discover trends and analyze information by providing the following features:

• Add, remove or reorder attributes, dimensions and measures
  
• Drill down display detail data or drill up to display summary data
  
• Pivot the crosstab , which swaps the colums and rows
  
• Sort data
  
• Filter data
  
• Display columns or rows that show aggregate data
  
• Create a column to display calculated data
  
• Choose from several chart types to display data
  
• Export content to other formats (PDF, Excel, Word etc)

You can download the sample BIRT Object used in the recording from here. Working with Actuate BIRT Analyzer requires Actuate iServer, download an evaluation copy of Actuate iServer and give it a try.

A typical BIRT report development process involves defining data sources, data sets/cubes and then using these data sets to create the required visualization. When the report is executed BIRT engine takes these definitions and fetches the data and renders it. One way to reuse these data definitions is to define them in a BIRT template or library and import it in your report development.

Starting with version 11 of Actuate, BIRT introduced a new construct that allows creation of data models as one or more flat /muti-dimensional structures from multiple and disparate data sources. This new data model referred to as "BIRT Data Objects" can be consumed by BIRT as a data source.



A BIRT Data Object shares the same "report design" perspective in BIRT Designer Professional and uses the similar concept of data sources, data sets and cubes that you are already familiar with. See Video


BIRT Data Object design files have the file extension "datadesign" (e.g. MyDataObject.datadesign) and once deployed in the iServer environment it can be scheduled to run in the background to generate highly compressed and optimized data object(e.g. MyDataObject.data) that contains a snapshot of sourced data at the time it was executed. This enables you to access data without spending time and resources data centralization activities. BIRT Data Objects can leverage a data warehouse or any data type that you have access to in BIRT. Inside a BIRT Object you can merge/join disparate data sets , apply data transformation and formatting routines and create computed columns based on your business logic.

BIRT Data Objects allows you to provide linking of a data set columns to other local or remote content allowing drilldown capability to operational dataset. For example, a link can be created from a “CustomerNumber” column of a dataset column to a “Customer Details” report.


If a BIRT Data Object contains multiple cubes you can share one or more dimensions across these cubes to facilitate simultaneous analysis for disparate measures along common dimensions. For example you can create a dashboard with “Sales revenue” and “Customer loyalty” as KPIs and then drill-down into these measures using the common “Geography” dimension.


BIRT Data Objects takes full advantage of 64 bit environment and avails of the increased memory address space for faster access and highly responsive ad-hoc queries and data analysis. Actuate’s in-memory architecture optimizes memory in a number of ways such as stripping data of redundancies, managing in-memory objects, sharing objects loaded into memory across a number of users and applications and providing support for 32-bit platform through intelligent disk-based caching

BIRT Data Objects models page level security (PLS) and gives you flexibility to secure the entire data object itself or rows & columns in data set and in case of cubes dimensions, measures & members.

The security comes in the form of Access Control Lists (ACL's) and lets you specify users/roles that can have access rights to the data. This eliminates the need to create multiple data objects to cater to the varying needs of consumers of this data.
All these features provides the foundation for supporting a wide spectrum of ad hoc query and reporting tools such as BIRT Studio, BIRT 360 and BIRT Data Analyzer that I will be talking about in upcoming blogs. Meanwhile you can download a free trial copy of Actuate version 11 products from here.

Ashwini Verma