ETL-processen : teori kontra praktik

Sammanfattning: Abstract Nowadays, companies and organizations manage large amounts of data to be used as support in decision making. A good solution for data management is in the form of data warehouses. A data warehouse is used to gather information from multiple sources within a company and arrange the information in a way that makes it easy to retrieve and analyze.   Designing a data warehouse is not a simple task. The study herein will focus on what a data warehouse is and how the ETL process is used for data storage in the data warehouse. ETL stands for Extract, Transform and Load and is the process that retrieves data, transforms it and finally loads it into the data warehouse. In addition to gathering information from literature we have also conducted an interview with data architect Rasmus Lindén with Know IT AB. During this interview Lindén described how Know IT AB had constructed a data warehouse for Upplands Lokaltrafik, with primary focus on how they used the ETL process to treat data to be stored in UL’s data warehouse.   The purpose of this paper is to examine the theory behind the ETL process and subsequently investigate how it may be applied by comparing the theory and how the company Know IT AB has applied it when designing data warehouses. One of the observations we made is that the theory we have studied fails to mention or illustrate some factors we during our interview found were considered important in designing data warehouses.