A team uses Databricks Asset Bundles to manage ML assets. They have a bundle that defines an MLflow experiment, a registered model, and a serving endpoint. They need to deploy the bundle to a new environment without affecting the existing environment. What is the best practice?

Manually create the resources in the new environment and then update the bundle to match.

Create a new bundle for the new environment and copy the configuration.

Use the same bundle and change the target environment in the deployment command.

Use a separate workspace for the new environment and redeploy the same bundle.

A team uses Databricks for ML development. They have a mix of long-running training jobs and short interactive queries. They want to optimize cost without sacrificing performance. What is the best compute strategy?

Use a fixed-size cluster for all workloads to ensure predictable performance.

Use a single cluster with autoscaling and a large maximum size.

Use a pool for all clusters and disable autoscaling.

Use separate clusters: one with autoscaling for interactive queries and one with autoscaling for training jobs.

Why is Delta Lake a recommended storage format for machine learning data?

It stores data in a proprietary format that is faster than Parquet.

It provides ACID transactions, time travel, and efficient upserts.

It is the only format that can be read by MLflow.

It automatically trains machine learning models.

A company stores ML training data in Delta Lake. They want to ensure that data is versioned and can be time-traveled for reproducibility. What is the best practice for managing this data?

Use external tables in Unity Catalog without Delta Lake.

Store data in Parquet files and maintain a separate versioning system.

Store data in CSV files and use a naming convention with timestamps.

Use Delta Lake and rely on its transaction log for versioning and time travel.

DatabricksCertified Machine Learning Professional

Domain 2Objective 3

Environment Architectures MACHINE-LEARNING-PROFESSIONAL Practice Questions (Page 1)

Part of the ML Ops domain, which accounts for 44% of the MACHINE-LEARNING-PROFESSIONAL exam. Databricks does not publish an official question count, but from its 120-minute exam (~50–80 total, ~22–35 in this domain), expect 4–7 from this objective — we provide 28 practice questions to prepare you well beyond it. (estimate)

28questions here

6free pages

9concepts

44%of the exam

Questions 1–5

1expert · hard
A team uses Databricks Asset Bundles to manage ML assets. They have a bundle that defines an MLflow experiment, a registered model, and a serving endpoint. They need to deploy the bundle to a new environment without affecting the existing environment. What is the best practice?
Select an answer first
2application · medium
A team uses Databricks for ML development. They have a mix of long-running training jobs and short interactive queries. They want to optimize cost without sacrificing performance. What is the best compute strategy?
Select an answer first
3foundation · easy
Why is Delta Lake a recommended storage format for machine learning data?
Select an answer first
4application · medium
A company stores ML training data in Delta Lake. They want to ensure that data is versioned and can be time-traveled for reproducibility. What is the best practice for managing this data?
Select an answer first
5foundation · easy
Which file format is used to define a Databricks Asset Bundle?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to MACHINE-LEARNING-PROFESSIONAL

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Databricks. “MACHINE-LEARNING-PROFESSIONAL” is a trademark of its owner, used for identification only.