DEA-C01 Data Operations and Support Practice Test 4 — 15 Questions

Question 1

A company runs an Amazon EMR cluster with Spark jobs. One job fails with 'Container killed by YARN for exceeding memory limits'. The data engineer has already increased the executor memory. What is the NEXT best step to resolve the issue?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Increase the YARN container memory allocation (yarn.nodemanager.resource.memory-mb).. When YARN kills a container for exceeding memory limits, it is usually because the total memory footprint (executor JVM heap + off-heap memory + overhead) exceeds the YARN container size. Increasing spark.executor.memory alone does not increase the YARN container size — the container is sized by yarn.nodemanager.resource.memory-mb and the scheduler's allocation. The next step is to increase the YARN container memory allocation so the executor has room for heap plus overhead.

Answer

Set spark.executor.memoryOverhead to a higher value.

Answer

Decrease the number of Spark partitions.

Answer

Increase the driver memory.

Question 2

Refer to the exhibit. A data engineer has attached this IAM policy to a user. The user reports being unable to upload files to my-bucket from an on-premises network with a public IP of 203.0.113.5. What is the issue?

Accepted Answer

The user's IP address is not within the allowed IP range. The IAM policy includes a condition that restricts access to requests originating from the IP range 10.0.0.0/16 (a private range). The user's on-premises network has a public IP of 203.0.113.5, which is not within that range, so the condition fails and the upload is denied. Option A is incorrect because the resource ARN does include the bucket itself (arn:aws:s3:::my-bucket/*), so that is not the issue. Option C is incorrect because the policy explicitly allows s3:PutObject. Option D is incorrect because there is no condition requiring server-side encryption.

Answer

The resource ARN does not include the bucket itself

Answer

The user does not have s3:PutObject permission

Answer

The bucket requires server-side encryption

Question 3

Refer to the exhibit. A data engineer runs this CloudWatch Logs Insights query on a log group but gets no results. What is the most likely reason?

Accepted Answer

There are no log events containing the string 'ERROR' in the log group. The query 'fields @timestamp, @message | filter @message like /ERROR/ | sort @timestamp desc | limit 20' is syntactically correct (so option A is wrong). The limit of 20 would still return results if matching log events existed (so B is wrong). Although CloudWatch Logs Insights defaults to the last 15 minutes, that alone doesn't explain zero results—if errors occurred in that window, they would appear; the lack of results suggests no 'ERROR' events exist in the scanned logs (so C is wrong). The most likely reason is that the log group contains no log events with the string 'ERROR', making option D correct.

Answer

The query syntax is invalid

Answer

The query limit of 20 is too low

Answer

The time range is not specified and defaults to the last 15 minutes

Question 4

A company runs a data pipeline that ingests streaming data via Amazon Kinesis Data Streams, processes it with an AWS Lambda function, and stores results in Amazon DynamoDB. The Lambda function sometimes fails due to 'ProvisionedThroughputExceededException' on the DynamoDB table. Which combination of steps should a data engineer take to resolve this issue?

Accepted Answer

Enable DynamoDB auto scaling and configure a dead-letter queue for the Lambda function.. Enabling DynamoDB auto scaling adjusts read/write capacity based on traffic, preventing ProvisionedThroughputExceededException. Configuring a dead-letter queue for the Lambda function captures failed events for later analysis and reprocessing, ensuring no data loss. Together, these steps address both the throughput issue and error handling.

Answer

Increase the Lambda function timeout and enable batch windows.

Answer

Increase the number of Kinesis shards to reduce Lambda invocations.

Answer

Increase Lambda reserved concurrency and disable retries.

Question 5

A data engineer needs to troubleshoot a failed AWS Glue job that reads from an Amazon RDS for MySQL database. The error log shows 'Communications link failure'. Which step should the engineer take FIRST?

Accepted Answer

Check that the security group associated with the Glue job allows outbound traffic to the RDS database.. 'Communications link failure' is a JDBC/MySQL error indicating the client cannot establish a TCP connection to the database — typically a network reachability problem, not authentication or schema. Since AWS Glue jobs run in an AWS-managed VPC (or a customer VPC via a connection), the first thing to verify is that the Glue job's security group permits outbound traffic to the RDS instance on port 3306. Checking network connectivity before credentials or schema is the correct troubleshooting order.

Answer

Increase the job timeout and retry count.

Answer

Verify that the database username and password are correct in the Glue connection.

Answer

Confirm that the table schema in MySQL matches the Glue Data Catalog.

Question 6

A company uses Amazon EMR to process large datasets stored in Amazon S3. The data engineer notices that EMR tasks are failing with 'DiskOutOfSpace' errors. The cluster uses m5.xlarge instances with 1 EBS volume of 64 GB. What is the MOST cost-effective solution to resolve this issue?

Accepted Answer

Increase the EBS storage volume size for each instance and use spot instances for task nodes.. Increasing the EBS volume size provides additional disk space per instance, directly resolving the disk out-of-space error. Using spot instances for task nodes reduces cost. Option A is incorrect because mixing on-demand and spot instances does not increase per-instance disk space. Option C is incorrect because switching to D2 instances is more expensive and may not be necessary. Option D is incorrect because adding more task instances distributes the workload but does not increase the disk space available to each instance, so individual tasks may still fail due to disk space.

Answer

Use a mix of on-demand and spot instances for core nodes.

Answer

Switch to D2 instances which have more instance store volume.

Answer

Increase the number of task instances to distribute the workload.

Question 7

A data engineer is optimizing an Amazon Redshift cluster that runs a nightly ETL workload. The engineer notices that the query performance degrades over the week and improves after a VACUUM operation. Which action should the engineer take to automate this maintenance and minimize impact on performance?

Accepted Answer

Schedule a VACUUM command using a query scheduler like Amazon EventBridge.. It enables automated scheduling of VACUUM using Amazon EventBridge or Redshift's query scheduler, ensuring regular maintenance without manual intervention. Option A is wrong because manual intervention only when performance degrades is not automated and allows performance to degrade unnecessarily. Option B is wrong because disabling auto vacuum and running manual VACUUM nightly is not as efficient or integrated as using a scheduler, and it may disrupt the ETL if not timed properly. Option D is wrong because dropping and recreating tables weekly is disruptive and loses data or requires complex re-creation logic, whereas VACUUM re-sorts data in place.

Answer

Run VACUUM manually only when performance degrades significantly.

Answer

Disable auto vacuum and run a manual VACUUM every night after the ETL.

Answer

Drop and recreate the tables weekly to avoid unsorted data.

Question 8

A company runs a daily batch process that reads data from Amazon S3, transforms it with AWS Glue, and loads it into Amazon Redshift. The process takes 6 hours, but the business requires completion within 4 hours. Which design change would MOST reduce runtime?

Accepted Answer

Load data directly from S3 to Redshift using COPY command, then transform in Redshift. The biggest runtime reduction comes from eliminating the Glue transformation step and loading raw data directly into Redshift with the COPY command, then transforming inside Redshift using SQL. COPY is a massively parallel, highly optimized bulk-load utility that leverages Redshift's MPP architecture, and Redshift's columnar engine performs transformations far faster than a general-purpose Spark ETL job for this workload. This removes the 6-hour Glue bottleneck entirely.

Answer

Increase the number of Glue workers

Answer

Use S3 Select to filter data before Glue

Answer

Switch to columnar storage in Redshift

Question 9

A company uses Amazon S3 to store raw data files. An AWS Glue crawler creates metadata in the Data Catalog. The data engineer discovers that the crawler is not detecting new partitions after new data is added to the S3 bucket. What is the MOST likely cause?

Accepted Answer

The IAM role used by the crawler does not have kms:Decrypt permission for the KMS key that encrypts the new partitions.. When new partitions are added to an S3 bucket encrypted with AWS KMS (SSE-KMS), the AWS Glue crawler needs the kms:Decrypt permission to read the object metadata and detect the new partitions. Without this permission, the crawler cannot list the objects or determine the partition structure, so it fails to update the Data Catalog. Option B is incorrect because the crawler's partition detection is not disabled by default; the 'Crawl all folders' setting is for crawling subfolders, not for partition detection. Option C is incorrect because AWS Glue crawlers can handle large numbers of objects, and there is no specific limit that would prevent partition detection. Option D is incorrect because S3 event notifications are not required for crawling; the crawler periodically scans the S3 location and does not rely on events to detect changes.

Answer

The crawler configuration has 'Crawl all folders' disabled.

Answer

The S3 bucket has too many objects, exceeding the crawler's limit.

Answer

The crawler does not have S3 event notifications enabled.

Question 10

A data engineer is troubleshooting an AWS Step Functions workflow that calls a Lambda function to process data. The workflow sometimes fails with a 'StateMachineExecutionLimitExceeded' error. What is the MOST likely cause?

Accepted Answer

Number of concurrent executions exceeds the account limit. The error 'StateMachineExecutionLimitExceeded' indicates that the number of concurrent executions of the state machine has exceeded the account-level limit. AWS Step Functions has a default limit on concurrent executions per account per region, and when this limit is reached, new executions fail with this error. The other options are not directly related to this specific error.

Answer

Execution time exceeds the maximum allowed duration

Answer

Lambda function memory limit exceeded

Answer

Lambda function concurrency limit reached

Question 11

A data engineer is monitoring an Amazon Redshift cluster using Amazon CloudWatch. The engineer notices that the 'WriteThroughput' metric is consistently below the provisioned IOPS for the cluster's EBS volumes. The query performance is slower than expected. Which action is MOST likely to improve write performance?

Accepted Answer

Upgrade to a larger node type with more CPU and memory.. If WriteThroughput is consistently below provisioned IOPS, the EBS volumes are not the bottleneck — the cluster is not generating enough write I/O to saturate them. That points to a compute-side constraint (CPU/memory) limiting query and write processing, so upgrading to a larger node type with more CPU and memory is the most likely fix. Adding IOPS would only help if the volumes were saturated.

Answer

Reduce the number of concurrent queries to the database.

Answer

Add sort keys to the tables to improve data distribution.

Answer

Increase the provisioned IOPS on the EBS volumes.

Question 12

A data engineer is designing a disaster recovery strategy for an Amazon RDS for PostgreSQL database that is used in a data pipeline. The database must have a Recovery Point Objective (RPO) of less than 1 minute and a Recovery Time Objective (RTO) of less than 5 minutes. Which TWO actions should the engineer take?

Accepted Answer

Enable automated backups with point-in-time recovery.. Option B is correct because automated backups with point-in-time recovery continuously capture transaction logs (WAL) for Amazon RDS for PostgreSQL, allowing restoration to any second within the retention period, which satisfies the RPO of less than 1 minute. Option C is correct because a Multi-AZ deployment maintains a synchronously replicated standby instance in another Availability Zone and automatically fails over in typically 60–120 seconds, meeting the RTO of less than 5 minutes. Option A is incorrect because manual snapshots are periodic and do not provide continuous point-in-time recovery, so they cannot guarantee an RPO under 1 minute. Option D is incorrect because a read replica in a different AZ is asynchronously replicated and is not an automatic failover target for disaster recovery. Option E is incorrect because Amazon Aurora Global Database applies to Aurora, not to Amazon RDS for PostgreSQL.

Answer

Take frequent manual snapshots and copy them to another Region.

Answer

Create a read replica in a different Availability Zone.

Answer

Use cross-Region replication with Amazon Aurora Global Database.

Question 13

A company uses AWS Lake Formation to manage access to data in S3. A data analyst reports being unable to query a table in Amazon Athena, receiving an 'Access Denied' error. The analyst has SELECT permission on the table in Lake Formation. What additional configuration is MOST likely causing the issue?

Accepted Answer

The IAM role used by Athena does not have S3 GetObject permission on the underlying data. Lake Formation grants table-level permissions through the Glue Data Catalog, but the underlying S3 data access still requires the query engine's IAM role to have s3:GetObject on the data location. Even with SELECT granted in Lake Formation, Athena's execution role must be able to read the actual objects, or the query returns Access Denied. This dual-permission model (Lake Formation + IAM/S3) is the most common cause of this symptom.

Answer

Athena does not have permission to access the Glue Data Catalog

Answer

The analyst does not have DESCRIBE permission

Answer

The table is not registered with Lake Formation

Question 14

A data engineer is troubleshooting an Amazon EMR cluster that has been running for several days. The cluster uses Amazon S3 as the data source and HDFS for intermediate storage. The engineer notices that some tasks fail with 'Java heap space' errors. Which TWO actions should the engineer take to resolve this issue?

Accepted Answer

Increase the maximum Java heap size for the task nodes (mapreduce.map.java.opts).. The 'Java heap space' error indicates that the JVM running the map/reduce task has run out of heap memory, so option C is correct: increasing the maximum Java heap size via mapreduce.map.java.opts (and correspondingly mapreduce.reduce.java.opts) directly gives the task JVM more heap to work with. Option D is also correct because yarn.nodemanager.resource.memory-mb controls the total physical memory YARN can allocate to containers on a node; raising it allows larger containers (and thus larger heaps) to be scheduled, which is necessary when the heap increase would otherwise exceed the node's advertised YARN memory. Option A (EMRFS consistent view) addresses S3 metadata consistency and listing performance, not JVM heap exhaustion. Option B (more containers per node) would reduce per-container memory and worsen heap pressure. Option E (decreasing container size) shrinks the memory available to each task, which would make the heap errors more frequent.

Answer

Enable EMRFS consistent view for S3.

Answer

Increase the number of containers per node.

Answer

Decrease the YARN container size.

Question 15

A company uses Amazon DynamoDB as the primary data store for a web application. The application experiences high read latency. Which TWO actions can improve read performance?

Accepted Answer

Enable DynamoDB Global Tables. Global Tables replicate data across multiple AWS Regions, allowing the application to read from a local replica with lower network latency instead of crossing regions to the primary table. Option C (Enable DynamoDB Accelerator (DAX)) is correct because DAX is an in-memory cache purpose-built for DynamoDB that serves eventually consistent reads in microseconds, dramatically reducing read latency for read-heavy workloads. Option A (Add a Global Secondary Index) does not inherently improve read performance; a GSI provides an alternative query path but consumes its own read capacity and can add latency if not designed well. Option D (Enable DynamoDB Streams) captures item-level changes for event-driven processing and does not speed up reads. Option E (Increase the write capacity units) only affects write throughput and has no direct impact on read latency.

Answer

Add a Global Secondary Index (GSI)

Answer

Enable DynamoDB Streams

Answer

Increase the write capacity units