Databricks Certified Data Engineer Associate Exam 2026:
Professional Certification Practice Test — Lakehouse Architecture,
Delta Lake, Data Ingestion, ETL Pipelines, Spark SQL & Workflow
Orchestration — 190 Multiple-Choice Questions with Answers &
Detailed Rationales
Section 1: Databricks Data Intelligence Platform (Questions 1–20)
Question 1
Which Databricks compute type is designed for running automated production workflows
with automatic restart and cost optimization?
A. All-purpose cluster
B. SQL warehouse
C. Job cluster
D. Serverless SQL warehouse
Correct Answer: C
Explanation: Job clusters are dedicated compute resources created for running automated
jobs, offering lower cost than all-purpose clusters and terminating after job completion.
Question 2
A data engineer needs to provide interactive analysis capabilities for a data science team
while ensuring isolation from production workloads. Which compute configuration is most
appropriate?
A. Single-node cluster
B. High-concurrency cluster
C. All-purpose cluster
D. Job cluster
,Correct Answer: C
Explanation: All-purpose clusters support interactive notebooks and collaboration, allowing
multiple users to share compute while keeping production jobs separate.
Question 3
Which Databricks compute service provides instant startup and automatic scaling without
manual cluster configuration?
A. Job cluster
B. Classic all-purpose cluster
C. Serverless compute
D. Single-node cluster
Correct Answer: C
Explanation: Serverless compute eliminates cluster management overhead, providing instant
availability and automatic scaling for notebooks and workflows.
Question 4
What is the primary advantage of using SQL warehouses in Databricks for BI workloads?
A. They support PySpark transformations
B. They are optimized for SQL analytics with Photon acceleration
C. They can execute Python UDFs natively
D. They provide unlimited concurrent query capacity
Correct Answer: B
,Explanation: SQL warehouses are purpose-built for SQL analytics, leveraging Photon engine
for high-performance query execution and BI tool integration.
Question 5
A team needs to run a machine learning training workload that requires GPU instances.
Which compute configuration supports this requirement?
A. SQL warehouse
B. Job cluster with GPU-enabled node types
C. Serverless SQL warehouse
D. Single-node CPU cluster
Correct Answer: B
Explanation: Job clusters can be configured with GPU-enabled instances and appropriate ML
runtimes for machine learning training workloads.
Question 6
Which factor most directly influences the cost of running an all-purpose cluster in
Databricks?
A. Number of notebooks attached
B. Number of DBUs consumed multiplied by cloud instance cost
C. Number of users in the workspace
D. Size of Unity Catalog metastore
Correct Answer: B
, Explanation: Databricks billing is based on Databricks Units (DBUs) consumed, which depend
on compute configuration and runtime, multiplied by cloud provider costs.
Question 7
What happens to a job cluster by default after the associated job completes?
A. It remains running for 24 hours
B. It is terminated automatically
C. It converts to an all-purpose cluster
D. It is archived but remains accessible
Correct Answer: B
Explanation: Job clusters terminate automatically upon job completion, minimizing costs for
production workloads.
Question 8
A data engineer wants to reduce cluster startup time for recurring jobs that require
consistent compute resources. Which feature should they configure?
A. Autoscaling
B. Cluster pools
C. Spot instances
D. Photon acceleration
Correct Answer: B
Explanation: Cluster pools maintain idle instances that can be quickly allocated to clusters,
reducing startup time for recurring workloads.
Professional Certification Practice Test — Lakehouse Architecture,
Delta Lake, Data Ingestion, ETL Pipelines, Spark SQL & Workflow
Orchestration — 190 Multiple-Choice Questions with Answers &
Detailed Rationales
Section 1: Databricks Data Intelligence Platform (Questions 1–20)
Question 1
Which Databricks compute type is designed for running automated production workflows
with automatic restart and cost optimization?
A. All-purpose cluster
B. SQL warehouse
C. Job cluster
D. Serverless SQL warehouse
Correct Answer: C
Explanation: Job clusters are dedicated compute resources created for running automated
jobs, offering lower cost than all-purpose clusters and terminating after job completion.
Question 2
A data engineer needs to provide interactive analysis capabilities for a data science team
while ensuring isolation from production workloads. Which compute configuration is most
appropriate?
A. Single-node cluster
B. High-concurrency cluster
C. All-purpose cluster
D. Job cluster
,Correct Answer: C
Explanation: All-purpose clusters support interactive notebooks and collaboration, allowing
multiple users to share compute while keeping production jobs separate.
Question 3
Which Databricks compute service provides instant startup and automatic scaling without
manual cluster configuration?
A. Job cluster
B. Classic all-purpose cluster
C. Serverless compute
D. Single-node cluster
Correct Answer: C
Explanation: Serverless compute eliminates cluster management overhead, providing instant
availability and automatic scaling for notebooks and workflows.
Question 4
What is the primary advantage of using SQL warehouses in Databricks for BI workloads?
A. They support PySpark transformations
B. They are optimized for SQL analytics with Photon acceleration
C. They can execute Python UDFs natively
D. They provide unlimited concurrent query capacity
Correct Answer: B
,Explanation: SQL warehouses are purpose-built for SQL analytics, leveraging Photon engine
for high-performance query execution and BI tool integration.
Question 5
A team needs to run a machine learning training workload that requires GPU instances.
Which compute configuration supports this requirement?
A. SQL warehouse
B. Job cluster with GPU-enabled node types
C. Serverless SQL warehouse
D. Single-node CPU cluster
Correct Answer: B
Explanation: Job clusters can be configured with GPU-enabled instances and appropriate ML
runtimes for machine learning training workloads.
Question 6
Which factor most directly influences the cost of running an all-purpose cluster in
Databricks?
A. Number of notebooks attached
B. Number of DBUs consumed multiplied by cloud instance cost
C. Number of users in the workspace
D. Size of Unity Catalog metastore
Correct Answer: B
, Explanation: Databricks billing is based on Databricks Units (DBUs) consumed, which depend
on compute configuration and runtime, multiplied by cloud provider costs.
Question 7
What happens to a job cluster by default after the associated job completes?
A. It remains running for 24 hours
B. It is terminated automatically
C. It converts to an all-purpose cluster
D. It is archived but remains accessible
Correct Answer: B
Explanation: Job clusters terminate automatically upon job completion, minimizing costs for
production workloads.
Question 8
A data engineer wants to reduce cluster startup time for recurring jobs that require
consistent compute resources. Which feature should they configure?
A. Autoscaling
B. Cluster pools
C. Spot instances
D. Photon acceleration
Correct Answer: B
Explanation: Cluster pools maintain idle instances that can be quickly allocated to clusters,
reducing startup time for recurring workloads.
