A team wants to use DLT for streaming inference but is unsure whether to use a streaming table or a materialized view. The source is a Delta table that is updated continuously. The team needs to apply a model to new data as it arrives. What should they use?

A materialized view with a trigger interval of 1 second to simulate streaming.

A streaming table only if the source is Kafka; otherwise, use a materialized view.

A materialized view, because it refreshes automatically when the source changes.

A streaming table, because it processes new data incrementally as it arrives.

When defining a DLT pipeline for streaming inference, what is the purpose of the 'target' setting in the pipeline configuration?

It specifies the streaming source to read from.

It specifies the cluster size for the pipeline.

It specifies the database where the output tables will be created.

It specifies the machine learning model to use for inference.

A streaming inference pipeline in DLT must compute a rolling average of sensor readings over a 10-minute window and then apply a model that requires this aggregated feature. The team notices that the pipeline restarts frequently due to cluster resizing, and predictions become incorrect after restarts. What is the most likely cause and the best fix?

The streaming table is defined as a view; change it to a materialized view to persist the state.

The model is not registered in MLflow; register the model to ensure it is loaded correctly after restarts.

The window duration is too long; reduce the window to 5 minutes to avoid state overflow.

The state store is not configured for fault tolerance; enable checkpointing and ensure the state store is durable.

A DLT pipeline for streaming inference is failing because the source Kafka topic has a schema that changed (a new column was added). The pipeline is using a schema that is no longer compatible. What is the best way to handle this in DLT?

Update the DLT pipeline code to reflect the new schema and restart the pipeline.

Recreate the Kafka topic to match the old schema.

Use schema evolution in the Delta table to automatically add the new column.

Set the pipeline to ignore schema changes by using a wildcard in the select statement.

A DLT pipeline for streaming inference is running but producing no output. The source is a Kafka topic that is receiving data. The pipeline is defined with a streaming table. What is the first thing to check?

Verify that the Kafka topic has data by consuming it manually.

Check the pipeline event log for errors or warnings.

Check the Delta table for new files.

Restart the pipeline to see if it recovers.

DatabricksCertified Machine Learning Associate

Domain 4Objective 4

Identify How Streaming Inference Is Performed with Delta Live Tables MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 4)

Part of the Section 4: Model Deployment domain, which makes up ~30% of our current practice bank. Databricks does not publish an official question count, but from its 90-minute exam (~35–60 total, ~11–18 in this domain), expect 2–3 from this objective — we provide 20 practice questions to prepare you well beyond it. (estimate)

20questions here

4free pages

5concepts

Questions 16–20

16application · medium
A team wants to use DLT for streaming inference but is unsure whether to use a streaming table or a materialized view. The source is a Delta table that is updated continuously. The team needs to apply a model to new data as it arrives. What should they use?
Select an answer first
17foundation · easy
When defining a DLT pipeline for streaming inference, what is the purpose of the 'target' setting in the pipeline configuration?
Select an answer first
18expert · hard
A streaming inference pipeline in DLT must compute a rolling average of sensor readings over a 10-minute window and then apply a model that requires this aggregated feature. The team notices that the pipeline restarts frequently due to cluster resizing, and predictions become incorrect after restarts. What is the most likely cause and the best fix?
Select an answer first
19application · medium
A DLT pipeline for streaming inference is failing because the source Kafka topic has a schema that changed (a new column was added). The pipeline is using a schema that is no longer compatible. What is the best way to handle this in DLT?
Select an answer first
20application · medium
A DLT pipeline for streaming inference is running but producing no output. The source is a Kafka topic that is receiving data. The pipeline is defined with a streaming table. What is the first thing to check?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

No more pagesBack to MACHINE-LEARNING-ASSOCIATE

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-ASSOCIATE” is a trademark of its owner, used for identification only.