This is the first of a three part series focused on an open source tool chain for small data business intelligence. Part 1 explores the history and reasoning behind dimensional modeling via a use case: online video viewing analysis. We examine the design of a star schema highlighting the utility of dimensional data structures. Specific topics include: date and time dimensions, data slicing, and choosing the appropriate grain for a fact table.

Part 2 will focus on Extract, Transform and Load with NodeJS while Part 3 will use gnuplot for visualizations. All three together represent a complete small data business intelligence solution from design to implementation to visualization.

Some Relevant Links

• Ralph Kimball Group has many resources for dimensional data modeling.
• Wistia for hosting videos plus the stats.
• Google's Timezone API for turning a lat, lon into an offset from UTC.
• And, as always, if you need an instance of PostgreSQL you could learn how to deploy one here. :)