Tuesday 15th November, 2011
9:30am to 12:30pm
Agile data warehousing can sound improbable. How can development teams proceed quickly by focusing upon a small set of requirements and not waste serious time on design mistakes? This course focuses on a key best practice for counterbalancing agile's inherent rush to deliver: semi-agile requirements management. Starting with the generic, agile notion of user stories for just-in-time requirements, we explore a more robust set of tools to manage requirements across many iterations and multiple release phases.
After an introduction to agile's five levels of planning, we will examine how to develop requirements from stakeholder requests, through use cases, all the way to their proper roles in system testing and user acceptance. The result will resolve many of the issues that often lead larger, more traditional BI/DW programs to categorically reject agile development methods.
We will examine templates and samples of the necessary artifacts for this more thorough approach, focusing on how they can help teams avoid design mistakes without slowing the project down. We will consider some of the features of a commercial tool for capturing business terms and evolving them into dimensions, measures, and star schemas while automating the process of enterprise data governance.
Prerequisite: This course builds upon Agile Data Warehousing 101.
You Will Learn
A framework for gathering requirements that uncovers unspoken business needs
A streamlined set of artifacts that provides definition and high-level control of agile BI/DW projects
How to use these artifacts to get new projects defined and funded
How agile user stories map to the more traditional notion of use cases
How to make use cases quicker to write and more appropriate for data warehousing
How to make a more direct connection between requirements and testing
also of interest to product owners and shared resources such as analysts, DBAs, testers, and technical managers
Sign in to add slides, notes or videos to this session