A Systematic Literature Review on Big Data Extraction, Transformation and Loading (ETL)

Data analytics plays a vital role in contemporary organizations, through analytics, organizations are able to derive knowledge and intelligence from data to support strategic decisions. An important step in data analytics is data integration, during which historic data is gathered from various sources and integrated into a centralized repository called data warehouse. Although there are various approaches for data integration, Extract Transform and Load (ETL) has become one of the most efficient and popular approach. Over the decades, ETL has been applied to a wide range of domains such as finance, health and telecom to mention but a few. As the popularity and use of ETL grow, it becomes important to analyze and identify the trends in the research and practice of ETL. In this paper, we perform a systematic literature review to identify and analyze: (1) Approaches used to implement existing ETL solutions (2) Quality attributes to be considered while adopting any ETL approach. (3) The depth of coverage in ETL research and practice with regards to the application domains, frequency publications and geographical locations of papers. (4) The prevailing challenges in developing ETL solutions. Furthermore, we discuss the implications of our findings to ETL researchers and practitioners.