Transaction and Analytic processing

Transaction Processing means allowing clients to make low-latency reads and writes, in contrast with batch processing that runs periodically. Usually a small number of records are looked up, inserted or updated with some key. This access pattern is also known as online transaction processing (OLTP).

A different access pattern is data analytics, which is an analytic query that needs to scan over a huge number of records, but only read a few columns per record to calculate aggregate statistics. This is usually called online analytic processing (OLAP). These operations can be huge and expensive.

Data Warehousing

A data warehouse is a separate database that analysts can query their content without affecting OLTP operations. The data warehouse contains a read-only copy of the data in all the various OLTP systems in the company.

Data warehouse is quite common for large companies as they have many different OLTP systems and a lot of data to process. The advantage of using a separate data warehouse, rather than querying OLTP systems directly is that the data warehouse can be optimized for analytic access.