hardMultiple ChoiceObjective-mapped

DP-203 Practice Question: Refer to the exhibit

Exhibit

Azure Databricks cluster configuration JSON:

{
  "cluster_name": "ETL Cluster",
  "spark_version": "10.4.x-scala2.12",
  "node_type_id": "Standard_DS3_v2",
  "autoscale": {
    "min_workers": 2,
    "max_workers": 8
  },
  "spark_conf": {
    "spark.sql.adaptive.enabled": "true",
    "spark.sql.adaptive.coalescePartitions.enabled": "true",
    "spark.sql.adaptive.advisoryPartitionSizeInBytes": "64MB"
  },
  "aws_attributes": {
    "first_on_demand": 1,
    "availability": "SPOT_WITH_FALLBACK",
    "zone_id": "us-west-2a"
  }
}

Refer to the exhibit. A data engineer notices that Spark jobs on this cluster are running slower than expected. The cluster is using spot instances with fallback. Which factor is most likely causing the performance degradation?

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

The cluster is using spot instances which may be frequently reclaimed

Spot instances can be preempted, causing delays. The configuration sets 'first_on_demand' to 1, meaning only 1 node is on-demand, and the rest are spot. Spot instances can be reclaimed, leading to recomputation and slower performance. The adaptive query execution settings are generally beneficial, not harmful.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
The spark.sql.adaptive settings are misconfigured
Why it's wrong here
These settings are standard and help performance; they are not likely the cause of slowness.
✓
The cluster is using spot instances which may be frequently reclaimed
Why this is correct
Spot instances are cheaper but can be terminated at any time, causing job failures or delays due to recomputation.
✗
The node type Standard_DS3_v2 is too small
Why it's wrong here
While node type affects performance, the question implies the cluster is running but slow; spot interruptions are a common issue.
✗
The autoscale configuration limits max workers to 8
Why it's wrong here
8 workers is generally sufficient; scaling is not the primary issue if the cluster is underutilized.

Go deeper

DP-203 Practice Question: Refer to the exhibit

Exhibit

Why each option matters

Option-by-option breakdown

Related to this question