Data modeling in Cassandra
The Sakila management wants us to migrate their database to Cassandra, they need to compare how Cassandra would compare against MongoDB. To this extent, your first task is to think of a data model for the new database.
When designing your data model, you’ll two follow two basic rules:
Your data model should distribute the data evenly across the nodes of the cluster.
For each frequent query, the number of partition to read should be as minimum as possible.
Exercise
Exercise 1 Analyze the Sakila business domain and propose a series of queries that the Sakila administration might want to run against the database.
Exercise
Exercise 2 Given the queries that you proposed in the first exercise, draw a workflow of the application that the Sakila management will use to query the data.
Exercise
Exercise 3 Based on the application workflow, draw the Chebotko diagram (the logical model) of the Sakila database.