Business intelligence: Performance and sustainability measures in an ETL process

Abstract

One of the critical factors for the success of any Business Intelligence (BI) solution is the performance of the extraction, transformation and loading process (Extract, Transform and Load — ETL). In reality, this phase is vital to ensure the timeliness, adaptability and quality of the information available to end users. However, despite the recognized importance attributed to the optimization of ETL, it is rare to find Information Technology (IT) teams that develop metrics and indicators that allow the factual monitoring of their performance. In line with the saying “the shoemaker's son always goes barefoot”, those responsible for the development of ETL processes often base their optimization exclusively on previous experiences and professional sensitivity. In this context, the objective of this article is to study the concept of BI for BI, analyze its added value in the optimization of ETL processes and indicate possible monitoring indicators.