“I know that you believe that you understand what you think I said, but I am not sure you realize that what you heard is not what I meant.”
-Sign on Pentagon office wall
1. Dimensionalize your schema.
2. Use Columnar techniques when appropriate.
3. Distribute your data by hash or replication with joins in mind.
4. Replicate frequently joined rows on Dimension Tables.
5. Use Logical partitioning on Fact tables when appropriate.
6. Make your Fact tables skinny.
7. Index your tables.
8. Consider denormalizing based on your environment.
Above are the eight rules for modeling Aster Data. These are designed around three principles ...