Building Data Warehouses Using Automation

Abstract

Software development is a complex endeavor. While significant benefits can be achieved, the process is often laborious, time consuming and error prone requiring multiple iterations in order to achieve the desired result. Issues arise for numerous reasons – coding defects, unclear requirements, migration challenges, lack of convention, and inadequate testing to name a few. When convention and automation are introduced into the software development lifecycle there are significantly fewer opportunities for failure. Automation also allows for shorter development windows. Generally there are fewer errors throughout testing, with the bulk of those being found in unit and functional testing, far before the users get involved in systems acceptance testing. A data warehouse consists of multiple subject areas in which many tasks are common and should be automated for the sake of efficiency and enforcing convention. This article discusses a set of tools that can be used to automate writing data warehouse objects. The article also provides statistics of time saved using automation.