data warehouse

Data Warehouse

A central database designed for analyzing data, not for running the apps that produce it.

When you'd see it: Anywhere analysts run SQL. Snowflake, BigQuery, Redshift, Databricks — all data warehouses (or close cousins). When someone says "pull this from the warehouse," they mean the analytical database where cleaned, joined, ready-to-query data lives.

Why it matters: Production databases are tuned for fast small writes — taking orders, recording clicks. Asking them to run a quarter-wide analytical query usually slows the app down or times out. The warehouse exists so the business can ask big questions without breaking the systems that run the business.

Common mistakes: Confusing the warehouse with the production database. The warehouse is downstream — a copy, refreshed on a schedule. Numbers in the warehouse can lag the live app by minutes or hours. Acting on warehouse data as if it's real-time is a common mistake that bites at month-end.

Related SQL, ETL, OLAP