Schema design is one of the areas which has often been overlooked yet can play a critical role when determining overall application performance when working with HBase. This class is based on practical experience and lessons learned on the importance of breaking away from the traditional relational-model approach to schema design, and is appropriate to all levels of individuals who are looking to improve their knowledge of HBase and of designing effective schemas.
We will cover the following:
- Trade-offs between a flattened schema vs. storing complex structures in cells
- The use of column families
- Different key designs, including hashing, salting and composite keys
- Secondary Indexing
It is assumed that you have a basic understanding of the basic fundamentals of HBase and Hadoop.