Developing Real Time Business Intelligence Systems the Agile Way

The paper cites why traditional waterfall methodologies fail to deliver in real-time business intelligence projects and how agile methodologies like Scrum can be adopted in their place to ensure success.

Abstract

Real-time business intelligence (Bl) systems are the buzz of the day. They harness the capabilities of real-time data warehouses to equip the users with current data and enable them to make correct decisions. A typical real time Bl system transforms and loads data into a data warehouse as soon it appears in the source systems. This is achieved using a complex network of Database, Web and Application servers coupled with ETL server, source systems and the intranet. Developing and maintaining these Bl systems is a challenging activity where the biggest challenge lies in making the systems work with almost zero latency to make it truly 'real-time'. The paper cites why traditional waterfall methodologies fail to deliver in such projects and how agile methodologies like Scrum can be adopted in their place to ensure success. The paper is supplemented with a live case study of an Infosys Technologies Limited (NASDAQ:INFY) project that uses Scrum and Global Delivery Model (GDM) to successfully manage a complex project of developing and maintaining a Bl system right from the requirements elicitation phase.