Summary
OLAP and data warehousing are both used in the context of specific database design patterns, where a database is used for analysis and reporting. OLAP stands for online analytical processing. In contrast to OLTP, OLAP implies bigger amounts of data, a smaller number of concurrent sessions and transactions, but the size of the transactions is bigger. The database structure is often denormalized to improve query performance. A database that is a part of an OLAP solution is often called a data warehouse.
In this chapter, we covered how to structure data in a data warehouse, how to load data there, and how to optimize the database performance by applying partitioning, using parallel query execution, and index-only scans.
In the next chapter, we will discuss the extended data types supported by PostgreSQL such as arrays, JSON, and others. These data types make it possible to implement complicated business logic using native database support.
