John Paul Ashenfelter, CTO, TransitionPoint
Track: Databases
Date: Monday, August 1st, 2005
Time: 1:30pm - 5:00pm
Location: Portland 253
Building a data warehouse is not rocket science, but it takes a different approach than traditional application development practices, even for experienced database developers. This tutorial focuses on building data warehouses using MySQL. We're going to dig into why MySQL specifically is a good choice for data warehousing, the basics of building an order-focused data warehouse, and the complementary open source tools for reporting and visualization. While there are MySQL databases in the terabyte range, we'll focus specficially on the 10-100 GB range.
Open source for data warehousing?
- Why build a data warehouse?
- Why use open source (MySQL)?
- Can this really work?
Basics of data warehousing
- Designing data warehouse tables and denormalization
- Facts and dimensions, and using MySQL to build them
- Building an order warehouse
- Specifics of data extraction and transform using a real order system
MySQL specifics
- tables, datatypes, and functions
- new in MySQL 5.0...
- performance considerations
Delivering the data warehouse
- open source reporting tools (particularly JasperReports)
- open source visualization tools
- delivery systems for data warehouse reports
We'll discuss other data warehouse issues as time and interest permit. By the end of the tutorial, you should have enough information to build a useful data warehouse using MySQL.