OLAP
OLAP systems run complex aggregations over large historical datasets to power reporting and multidimensional analysis.
OLAP, or Online Analytical Processing, describes systems designed to answer complex analytical questions over large datasets. Where OLTP runs the operations of a business, OLAP helps people understand it — through reports, dashboards, and ad hoc exploration.
How It Works
OLAP queries typically scan and aggregate millions or billions of rows: total sales by region by quarter, year-over-year growth, customer cohorts. To make this efficient, OLAP systems use:
- Columnar storage, which reads only the columns a query needs and compresses well.
- Denormalized or star/snowflake schemas with fact and dimension tables.
- Pre-aggregation and cubes that precompute summaries across dimensions.
- Massively parallel processing (MPP) to spread work across nodes.
Classic OLAP cubes (e.g., from SQL Server Analysis Services) precompute multidimensional aggregates. Modern cloud data warehouses such as Snowflake, BigQuery, Amazon Redshift, and ClickHouse provide OLAP at scale without explicit cubes.
Why It Matters
OLAP enables data-driven decision-making by making large-scale analysis fast and interactive. Analysts can slice, dice, drill down, and roll up data along dimensions like time, geography, and product.
OLAP is deliberately unsuited to high-frequency single-row writes; it favors bulk loads and read-heavy scans. Data usually arrives from OLTP systems through ETL or ELT pipelines.
Related Terms
OLAP is the analytical counterpart to OLTP, lives in a data warehouse, and relies on columnar storage and denormalization.