datastore-selection

Decide where data should live: a relational OLTP database, a columnar warehouse, a document store, a key-value cache, a graph database, a data lake or lakehouse, or a combination. Use whenever someone asks which database or storage to use for an application, feature or data product; SQL or NoSQL; Postgres vs MongoDB vs Cassandra vs Redis vs Neo4j vs ClickHouse vs BigQuery; data lake vs warehouse vs lakehouse or Delta Lake; OLTP vs OLAP; whether a system needs ACID transactions or can live with eventual consistency; how the CAP theorem applies; or how to design polyglot persistence for a platform with very different kinds of data. Also use to review a storage choice that is struggling with scale, latency or cost. Not for picking distribution or partition keys inside a chosen warehouse, and not for pipeline design or data quality rules.