Databricks SQL concepts

This page introduces the core concepts you need to use Databricks SQL effectively.

Interfaces and tools

Interface	Description
SQL editor	A browser-based interface for writing and running SQL queries. Includes integrated AI assistance, code comments, and version history.
AI/BI	A business intelligence solution that includes dashboards and Genie spaces for self-service data analysis and conversational data exploration.
Notebooks	Interactive documents that combine SQL with Python, Scala, or R. Attach a notebook to a SQL warehouse to run SQL alongside other languages.
Metric views	A semantic layer for defining reusable business metrics with consistent calculations across queries and dashboards.
Alerts	Automated monitors that evaluate query results against a condition and deliver notifications when the condition is met.
Jobs	Scheduled execution of SQL queries for automated data processing and reporting workflows.
ETL	Define and refresh streaming tables and materialized views in Databricks SQL for incremental ETL pipelines.
REST API	A programmatic interface for automating and managing Databricks SQL objects.

Queries and computation

Concept	Description
Query	A SQL statement that retrieves or transforms data. Queries in Databricks SQL are saved, versioned, and reusable.
SQL warehouse	The compute resource that executes SQL queries. All Databricks SQL interfaces run queries on a SQL warehouse.
Query history	A record of executed queries including execution time, resource usage, and performance characteristics.
Query profile	A detailed view of a query's execution plan. Use it to identify bottlenecks and optimization opportunities.
Visualization	A graphical representation of query results, such as a bar chart, line chart, or table.

Data management

Concept	Description
Dashboard	A collection of visualizations and commentary built from one or more queries. Dashboards are shareable and support interactive filtering.
Metric view	A named, reusable definition of a business metric backed by a SQL expression. Enforces consistent calculations across teams and tools.
Streaming table	A table that ingests data incrementally from a streaming source, managed by a Databricks SQL pipeline.
Materialized view	A precomputed view whose results are stored and refreshed automatically by a Databricks SQL pipeline.

Authentication and authorization

Concept	Description
User and group	A user is a unique individual with access to the Databricks workspace. A group is a collection of users managed together for access control purposes.
Personal access token	A credential used to authenticate to the REST API and to connect third-party tools to SQL warehouses.
Access control list	A set of permissions attached to an object that controls which principals can perform which actions on it.
Unity Catalog	The governance layer for Databricks. Provides centralized access control, auditing, lineage, and data discovery across workspaces.