Data Virtualization for Agile Business Intelligence Systems 2012

In addition, new forms of reporting and analytics are being requested by the user community, such as operational analytics, 360˚ reporting, exploratory analysis, deep and big data analytics, self-service BI, and semi-structured and unstructured data analytics.

All these new requirements demand that BI systems are developed in a more agile way. One of the technologies making this possible today is data virtualization. In a nutshell, data virtualization decouples data sources from the application and reports, and by doing that it can present a heterogeneous set of data stores as one logical database to all the reports. Compared to ETL, where data integration takes place in a scheduled manner, with data virtualization data is integrated on-demand.

The last few years, various data virtualization servers have become available to develop these systems, including those of Composite, Denodo, IBM, Informatica, Information Builders, Queplix and RedHat. In many projects they have already proven that data virtualization technology is mature, does simplify BI systems, and makes them more severely more agile.