Design logical model of Datawarehouse, Fact Tables and Dimensions Table

Question

Hi i\'m newbie in Datawarehousing,For homework ask me realize the logical design, physical and implementation.How would you model this in a Data Warehouse:

i wish design

Answer 1

The first step in designing any data warehouse is to choose a business process. You have already done so by identifying reporting requirements based on baseball statistics, which is already a very well defined process (based on the rules of the game).

The second step is to identify the grain. The grain is the lowest level of detail you need to represent in reporting queries. For your baseball game situation, this would most likely be a pitch.

Next, you would identify the dimensions required to describe your grain. These are all easily identifiable - game, pitcher, batter and played date are obvious ones to start with.

Finally, you would identify the facts that hold measures relating to these dimensions. This encompasses many of the measures from your question, including whether a run was scored - which would be additive across any combination of dimensions, i.e. at inning, game, team, player, or season levels.