The promise of business intelligence (BI) requires that executives, managers, and systems engineers consider carefully what constitutes a complete and effective enterprise BI stack. Paul Pambudi (2010) describes a full BI stack as shown below:
Note that the data sources for BI include enterprise resource planning (ERP) systems, client resource management (CRM) systems, human resource management (HRM) systems, and others. Raw data from these sources is extracted, transformed, and loaded (ETL) into a data warehouse for storage. From the data warehouse, data is translated into usable information via a semantics layer, which serves the information to business analysts seeking to produce actionable descriptive, predictive, and prescriptive BI that adds value to decision-making.
The above illustration provides an instructive starting point for discussing the processes by which enterprise BI is produced. I will be discussing each of the layers depicted above in greater detail in coming posts.
Source: Pambudi, P (2010, October 10), Full BI stack is Not Only for Large Enterprises, Optimeon.
Whom to Listen to in the Fed Minutes
50 minutes ago