[CockroachDB Internals #1] Let’s talk about OLTP and OLAP databases and understand where they differ … ⚡
OLTP databases are designed to handle Transactional workloads. The word transactions do not mean ACID properties, instead, it denotes a specific type of data access pattern
- reads and writes on a small amount of data/rows/records
- the database will be queried concurrently by many clients
- read and write operation cannot run for long
On the other hand, OLAP databases handle the Analytical workloads that serve the following data access pattern
- reads involve accessing large number of data/rows/records
- writes can be granular or batch, but are mostly done in bulk
- bandwidth is more important than latency, which means read operations can run for long
Hope this clears up the difference between OLTP and OLAP. Some examples of OLTP are MySQL, Postgres, Redis, etc. While BigQuery and Redshift are OLAP databases.
Recently, I started diving deep into CockroachDB internals, and I will be sharing my learning in public. So, if you find it amusing, follow along.