1
AWS Certified Data
Engineer – Associate
(DEA-C01): Advanced
Practice Exam V2.0 a
well detailed one 2025 /
2026 written and graded
A+ upgraded
, 2
Title
AWS Certified Data Engineer – Associate (DEA-C01): Version 2.0 Comprehensive Advanced
Practice Exam with 150 High-Difficulty Scenario-Based Questions Covering Data Ingestion,
Transformation, Storage Management, Operations, Security, and Governance for Experienced
Cloud Data Professionals
Exam Domains Coverage
This question set covers all four domains of the DEA-C01 exam:
• Domain 1: Data Ingestion and Transformation (34% of scored content)
• Domain 2: Data Store Management (26% of scored content)
• Domain 3: Data Operations and Support (22% of scored content)
• Domain 4: Data Security and Governance (18% of scored content)
Questions 1–150
Domain 1: Data Ingestion and Transformation (Questions 1–51)
Question 1
A financial trading firm needs to ingest 500,000 stock trade events per second with sub-50ms
latency. The solution must support exactly-once processing, allow replay of data for 14 days,
and automatically scale during market volatility. Which architecture should the data engineer
implement?
A. Amazon Kinesis Data Streams with on-demand capacity mode and AWS Lambda with
idempotent processing
B. Amazon MSK with provisioned throughput and Kafka Streams with exactly-once semantics
C. Amazon Kinesis Data Streams with provisioned shards and KCL with DynamoDB checkpointing
D. Amazon SQS FIFO queues with EC2 Auto Scaling consumers
- detailed answer 100% correct :- A
, 3
Rationale: Kinesis Data Streams on-demand capacity mode automatically scales to handle traffic
spikes. Lambda provides sub-second processing with idempotent logic for exactly-once
semantics. 14-day retention is supported. MSK requires more management. SQS FIFO has
throughput limitations (300 messages per second).
Question 2
A data engineer is building an AWS Glue ETL job using Glue version 5.0. The job needs to
process data from multiple open table formats including Apache Iceberg, Delta Lake, and
Apache Hudi. Which configuration enables support for all three formats?
A. Use --extra-jars to include format-specific libraries
B. Use --spark.sql.catalogImplementation set to in-memory
C. Glue 5.0 includes native support for Iceberg, Delta Lake, and Hudi
D. Use AWS Glue Data Catalog with format-specific classifiers
- detailed answer 100% correct :- C
Rationale: AWS Glue 5.0 includes updated support for open table formats including Apache
Hudi 1.0.2, Apache Iceberg 1.10.0, and Delta Lake 3.3.2. No additional configuration is required
for these formats. Glue 5.0 also upgrades the core engine to Apache Spark 3.5.6 and Python
3.11.
Question 3
A company needs to migrate 200 TB of data from an on-premises Hadoop HDFS cluster to
Amazon S3. The network bandwidth is 500 Mbps, and the migration must complete within 10
days. Which approach should the data engineer recommend?
A. AWS DataSync with multiple task executions and bandwidth limit
B. AWS Snowball Edge with cluster deployment (multiple devices)
C. AWS Database Migration Service (DMS) with HDFS as source
D. AWS Transfer Family with SFTP and parallel transfers
- detailed answer 100% correct :- B
Rationale: At 500 Mbps, transferring 200 TB would take approximately 37 days, exceeding the
10-day requirement. Snowball Edge devices provide petabyte-scale physical data transport.
DataSync would be too slow. DMS is for databases, not HDFS. Transfer Family is for ongoing file
transfer.
, 4
Question 4
A data engineer is designing a streaming pipeline using Amazon Kinesis Data Streams with the
new On-demand Advantage mode. The stream must handle instant throughput increases up to
10 GiB/s during flash sales. Which Kinesis feature enables this capability?
A. Enhanced fan-out with HTTP/2
B. Warm throughput feature in On-demand Advantage mode
C. Provisioned shards with auto-scaling
D. Kinesis Data Analytics with parallel processing
- detailed answer 100% correct :- B
Rationale: The warm throughput feature in Kinesis On-demand Advantage mode allows
configuring on-demand streams with instantly available throughput capacity up to 10 GiB/s.
Enhanced fan-out is for consumers. Provisioned shards require manual scaling. Kinesis Data
Analytics is for processing, not ingestion capacity.
Question 5
An AWS Glue job is processing 500 GB of data with complex transformations and multiple joins.
The job is failing with executor memory errors. The engineer has already increased the number
of workers. Which optimization should the engineer implement next?
A. Enable --spark.sql.adaptive.coalescePartitions.enabled with --spark.sql.adaptive.enabled
B. Use --job-bookmark-option with job-bookmark-enable
C. Increase --spark.sql.shuffle.partitions to 2000
D. Switch to G.025X worker type for better memory management
- detailed answer 100% correct :- A
Rationale: Adaptive Query Execution (AQE) in Spark 3.x dynamically optimizes query plans and
coalesces partitions based on runtime statistics. This reduces memory pressure by optimizing
shuffles. Glue 5.0 with Spark 3.5.6 provides enhanced AQE capabilities. Bookmarks are for
incremental processing. Increasing shuffle partitions may worsen memory issues. G.025X has
less memory.
Question 6
AWS Certified Data
Engineer – Associate
(DEA-C01): Advanced
Practice Exam V2.0 a
well detailed one 2025 /
2026 written and graded
A+ upgraded
, 2
Title
AWS Certified Data Engineer – Associate (DEA-C01): Version 2.0 Comprehensive Advanced
Practice Exam with 150 High-Difficulty Scenario-Based Questions Covering Data Ingestion,
Transformation, Storage Management, Operations, Security, and Governance for Experienced
Cloud Data Professionals
Exam Domains Coverage
This question set covers all four domains of the DEA-C01 exam:
• Domain 1: Data Ingestion and Transformation (34% of scored content)
• Domain 2: Data Store Management (26% of scored content)
• Domain 3: Data Operations and Support (22% of scored content)
• Domain 4: Data Security and Governance (18% of scored content)
Questions 1–150
Domain 1: Data Ingestion and Transformation (Questions 1–51)
Question 1
A financial trading firm needs to ingest 500,000 stock trade events per second with sub-50ms
latency. The solution must support exactly-once processing, allow replay of data for 14 days,
and automatically scale during market volatility. Which architecture should the data engineer
implement?
A. Amazon Kinesis Data Streams with on-demand capacity mode and AWS Lambda with
idempotent processing
B. Amazon MSK with provisioned throughput and Kafka Streams with exactly-once semantics
C. Amazon Kinesis Data Streams with provisioned shards and KCL with DynamoDB checkpointing
D. Amazon SQS FIFO queues with EC2 Auto Scaling consumers
- detailed answer 100% correct :- A
, 3
Rationale: Kinesis Data Streams on-demand capacity mode automatically scales to handle traffic
spikes. Lambda provides sub-second processing with idempotent logic for exactly-once
semantics. 14-day retention is supported. MSK requires more management. SQS FIFO has
throughput limitations (300 messages per second).
Question 2
A data engineer is building an AWS Glue ETL job using Glue version 5.0. The job needs to
process data from multiple open table formats including Apache Iceberg, Delta Lake, and
Apache Hudi. Which configuration enables support for all three formats?
A. Use --extra-jars to include format-specific libraries
B. Use --spark.sql.catalogImplementation set to in-memory
C. Glue 5.0 includes native support for Iceberg, Delta Lake, and Hudi
D. Use AWS Glue Data Catalog with format-specific classifiers
- detailed answer 100% correct :- C
Rationale: AWS Glue 5.0 includes updated support for open table formats including Apache
Hudi 1.0.2, Apache Iceberg 1.10.0, and Delta Lake 3.3.2. No additional configuration is required
for these formats. Glue 5.0 also upgrades the core engine to Apache Spark 3.5.6 and Python
3.11.
Question 3
A company needs to migrate 200 TB of data from an on-premises Hadoop HDFS cluster to
Amazon S3. The network bandwidth is 500 Mbps, and the migration must complete within 10
days. Which approach should the data engineer recommend?
A. AWS DataSync with multiple task executions and bandwidth limit
B. AWS Snowball Edge with cluster deployment (multiple devices)
C. AWS Database Migration Service (DMS) with HDFS as source
D. AWS Transfer Family with SFTP and parallel transfers
- detailed answer 100% correct :- B
Rationale: At 500 Mbps, transferring 200 TB would take approximately 37 days, exceeding the
10-day requirement. Snowball Edge devices provide petabyte-scale physical data transport.
DataSync would be too slow. DMS is for databases, not HDFS. Transfer Family is for ongoing file
transfer.
, 4
Question 4
A data engineer is designing a streaming pipeline using Amazon Kinesis Data Streams with the
new On-demand Advantage mode. The stream must handle instant throughput increases up to
10 GiB/s during flash sales. Which Kinesis feature enables this capability?
A. Enhanced fan-out with HTTP/2
B. Warm throughput feature in On-demand Advantage mode
C. Provisioned shards with auto-scaling
D. Kinesis Data Analytics with parallel processing
- detailed answer 100% correct :- B
Rationale: The warm throughput feature in Kinesis On-demand Advantage mode allows
configuring on-demand streams with instantly available throughput capacity up to 10 GiB/s.
Enhanced fan-out is for consumers. Provisioned shards require manual scaling. Kinesis Data
Analytics is for processing, not ingestion capacity.
Question 5
An AWS Glue job is processing 500 GB of data with complex transformations and multiple joins.
The job is failing with executor memory errors. The engineer has already increased the number
of workers. Which optimization should the engineer implement next?
A. Enable --spark.sql.adaptive.coalescePartitions.enabled with --spark.sql.adaptive.enabled
B. Use --job-bookmark-option with job-bookmark-enable
C. Increase --spark.sql.shuffle.partitions to 2000
D. Switch to G.025X worker type for better memory management
- detailed answer 100% correct :- A
Rationale: Adaptive Query Execution (AQE) in Spark 3.x dynamically optimizes query plans and
coalesces partitions based on runtime statistics. This reduces memory pressure by optimizing
shuffles. Glue 5.0 with Spark 3.5.6 provides enhanced AQE capabilities. Bookmarks are for
incremental processing. Increasing shuffle partitions may worsen memory issues. G.025X has
less memory.
Question 6
