Databricks DA-ASSOC Glossary: Unity Catalog, Dashboards, and SQL Terms

Databricks DA-ASSOC glossary of SQL, dashboards, Genie, modeling, and data security terms.

Use this glossary when Databricks SQL and analytics terms start to blur together. Keep it beside the cheat sheet and resources, not as a substitute for them.

High-yield terms

Term Short meaning Why it matters on DA-ASSOC
Databricks Data Intelligence Platform Databricks platform layer for data, analytics, AI, governance, and SQL workloads Top-level platform scope on the exam
Unity Catalog Databricks governance layer for data objects, permissions, and lineage Core trusted-analytics boundary
SQL Warehouse Compute service used to run Databricks SQL queries Core execution and workflow term
Query history Record of executed SQL queries and their execution details Important for debugging and analysis
Query Profile Detailed execution view for a query Important for performance-analysis questions
Query Insights Databricks feature for spotting poorly performing queries Common performance-analysis aid
Photon Databricks execution engine for fast analytics workloads Common query-performance concept
Liquid clustering Table-layout optimization approach for filtered analytics workloads Optimization concept in query analysis
Materialized view Stored precomputed query result refreshed by the platform Commonly contrasted with streaming tables and standard views
Streaming table Table designed for continuously arriving data workflows Important in freshness and workload-shape questions
Genie space AI/BI workspace that helps users interact with curated data through natural language Distinct consumption and trust boundary
Trusted asset Curated query or object treated as reliable inside Genie workflows Important Genie governance concept
Certified dataset Data asset marked as trusted and governed for use Trusted-data concept
Delta Sharing Sharing data with external systems without raw-copy handoff Common external-consumption concept
Lakehouse Federation Querying external systems from Databricks without first ingesting all data Important import-versus-query decision concept
Auto Loader Incremental ingestion feature for files arriving in cloud storage Important import-path concept
Row grain What one row is intended to represent in a query result One of the most important correctness concepts
Widget parameter Dashboard or query input control used to filter or alter behavior Common dashboard question term
Alert threshold Condition that triggers a notification Important dashboard and alerting concept
Catalog Explorer UI for discovering and managing governed data assets Important Unity Catalog workflow term
External table Table that points to data outside Databricks-managed storage Commonly contrasted with managed tables
Managed table Table with more platform-managed storage lifecycle Governance and storage-boundary term

Commonly confused pairs

Pair Keep this distinction clear
inner join vs left join matched rows only versus preserve all rows from the left side
UNION vs UNION ALL deduplicate combined rows versus keep all rows
materialized view vs streaming table precomputed query result versus continuous data processing table
managed table vs external table platform-managed storage versus externally referenced storage
dashboard refresh vs alert trigger refresh the visual data versus notify on a condition
query history vs query profile execution record versus detailed execution breakdown
query profile vs query insights one-query execution detail versus broader performance insight tooling
certified dataset vs trusted asset governed trusted data object versus trusted query or object inside Genie
widget parameter vs dashboard filter logic input control value versus how the query actually uses it
Delta Sharing vs Lakehouse Federation governed external sharing versus querying external systems in place

If three terms blur together

Cluster Fast separation
row grain / join logic / aggregation what one row means, how rows combine, and how rows collapse into metrics
SQL Warehouse / Query History / Query Profile run the query, inspect what ran, and inspect why it behaved that way
materialized view / streaming table / standard query output precomputed refreshed result, continuous data table, or direct query result
dashboard / alert / Genie space visual consumption, threshold notification, or AI-assisted interaction layer
Unity Catalog / certified dataset / Delta Sharing governance layer, trusted data object, or external sharing mechanism
Auto Loader / Lakehouse Federation / workspace upload incremental ingestion, query-in-place access, or small manual upload path

One-sentence memory hooks

  • If the metric is wrong, think row grain before chart settings.
  • If the query is slow, think Query History, Query Profile, and Query Insights before random SQL edits.
  • If the result should be trusted by many users, think certified dataset and lineage.
  • If the question is about external consumption, think Delta Sharing.
  • If the question is about interactive natural-language analytics, think Genie space, not a dashboard alone.

Operational clusters worth keeping straight

Cluster What it usually signals on the exam
row grain / joins / windows SQL correctness questions
warehouse / history / profile / Photon query-execution and analysis questions
dashboards / alerts / parameters analytics-consumption questions
Genie / trusted assets / certified datasets AI/BI and trust-boundary questions
Unity Catalog / managed tables / external tables / Delta Sharing governance and sharing questions
Auto Loader / Marketplace / Federation data-import and data-access questions

If the confusion is really about…

Topic family Best page to revisit
SQL correctness and quick rules Cheat Sheet
official Databricks facts and docs Resources
pacing and review order Study Plan
overall exam framing Guide root
Revised on Sunday, May 10, 2026