DEA-C01 Data Operations and Support Practice Test 3 — 15 Questions

Question 1

A company uses AWS Database Migration Service (DMS) to migrate an on-premises Oracle database to Amazon RDS for PostgreSQL. The migration completes successfully, but the data engineer notices that some tables have fewer rows in the target than the source. Which DMS setting should be checked to ensure full data migration?

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

Accepted Answer

The task logs show that some rows failed to apply due to data type conversion errors.. If some rows failed to apply due to data type conversion errors, those rows would be logged as errors and not written to the target, resulting in fewer rows. AWS DMS task logs capture these failures, and checking them is the direct way to identify rows that were skipped or rejected during migration. This is the most common cause of row count mismatches after a successful DMS task.

Answer

The LOB mode is set to 'Limited LOB mode' instead of 'Full LOB mode'.

Answer

The 'Enable validation' option is turned off.

Answer

The 'Parallel Apply' feature is disabled, slowing down the migration.

Question 2

A data engineer needs to transfer 50 TB of data from an on-premises HDFS cluster to Amazon S3. The on-premises network has a 1 Gbps link to AWS. Which AWS service should be used to perform the transfer efficiently?

Accepted Answer

AWS DataSync. AWS DataSync is the correct choice because it is designed for large-scale data transfers over the network, handling 50 TB efficiently over a 1 Gbps link by parallelizing transfers and automating scheduling. Amazon S3 Transfer Acceleration only improves upload speed over long distances but is not a dedicated transfer service. AWS Snowball Edge would be slower due to shipping and is unnecessary since the network link is sufficient. AWS Direct Connect provides a dedicated network connection but is not a data transfer service.

Answer

Amazon S3 Transfer Acceleration

Answer

AWS Snowball Edge

Answer

AWS Direct Connect

Question 3

A company runs a nightly Amazon EMR job that processes data from S3 and writes results back to S3. The job fails with 'OutOfMemoryError' in the reduce phase. The cluster currently uses 5 m5.xlarge instances. Which cost-effective change should the data engineer make?

Accepted Answer

Increase the number of reducers (mapreduce.reduce.tasks) and keep the same instance type.. Increasing the number of reducers (mapreduce.reduce.tasks) distributes the memory load across the existing 5 m5.xlarge instances, reducing the per-reducer memory pressure and preventing OutOfMemoryError without adding cost. Option A (adding more core nodes) increases cost without directly addressing reducer memory. Option C (reducing input data) may affect completeness and is not a proper fix. Option D (switching to r5.xlarge) is more expensive per instance, making it less cost-effective.

Answer

Increase the number of core nodes to 10.

Answer

Reduce the input data size by filtering early in the job.

Answer

Switch to r5.xlarge instances for more memory per instance.

Question 4

A company is running a critical data pipeline using AWS Glue. The pipeline must be highly available and fault-tolerant. Which TWO strategies should the data engineer implement? (Choose TWO.)

Accepted Answer

Configure the Glue job to run in multiple Availability Zones.. Option A is correct because AWS Glue jobs run on a managed Spark environment distributed across multiple Availability Zones by default, and explicitly configuring the job to run in multiple AZs ensures that the pipeline continues processing even if one AZ experiences an outage, which is essential for high availability and fault tolerance. Option D is correct because enabling job retries with exponential backoff allows transient failures (such as throttling, network issues, or temporary service disruptions) to be automatically retried without manual intervention, improving fault tolerance and resilience of the critical data pipeline. Option B is incorrect because using a single instance type for all Glue workers does not improve availability or fault tolerance; it may actually reduce flexibility and performance tuning options. Option C is incorrect because increasing concurrent runs affects throughput and parallelism, not fault tolerance, and could even cause resource contention or duplicate processing. Option E is incorrect because disabling job bookmarks would cause reprocessing of already-processed data, which harms reliability and efficiency rather than improving fault tolerance.

Answer

Use a single instance type for all Glue workers.

Answer

Increase the number of concurrent runs for the Glue job.

Answer

Disable job bookmarks to avoid reprocessing.

Question 5

A data pipeline uses AWS Step Functions to orchestrate multiple Lambda functions for data transformation. The pipeline occasionally fails with a 'StateMachineExecutionLimitExceeded' error. What is the MOST likely cause?

Accepted Answer

The account has reached the maximum number of concurrent state machine executions.. The error 'StateMachineExecutionLimitExceeded' specifically indicates that the AWS account has hit the maximum number of concurrent Step Functions state machine executions, which is a service quota (default 1,000,000 open executions per account, but regional limits and per-state-machine limits also apply). This is not a Lambda concurrency issue or an API Gateway rate limit; it is a Step Functions execution quota. The fix is typically to request a quota increase or reduce concurrent executions.

Answer

The API Gateway endpoint used by Step Functions has a rate limit.

Answer

The Lambda functions have reached their concurrent execution limit.

Answer

The state machine definition has a syntax error causing infinite loops.

Question 6

A data engineer needs to monitor the number of Amazon S3 PUT requests that result in a 403 AccessDenied error. Which AWS service should be used to capture the necessary data?

Accepted Answer

AWS CloudTrail data events. S3 request metrics do not provide filtering by status code; the 4xxErrors metric aggregates all 4xx errors. To monitor specific 403 AccessDenied errors, you should use AWS CloudTrail data events, which capture API call details including error codes. You can create a CloudWatch metric filter on the CloudTrail log group to emit a custom metric for 403 errors, enabling monitoring and alarming.

Answer

Amazon CloudWatch S3 request metrics

Answer

Amazon S3 server access logs

Answer

AWS Config

Question 7

Refer to the exhibit. A data engineer is troubleshooting an IAM policy attached to a user who cannot list objects in the S3 bucket 'example-bucket'. What is the most likely reason?

Accepted Answer

The bucket policy explicitly denies access to the user.. The question asks why a user cannot list objects in 'example-bucket'. The correct answer is that the bucket policy explicitly denies access. In AWS IAM, an explicit Deny in any policy (identity-based or resource-based) overrides any Allow. Even if the user's IAM policy grants s3:ListBucket, a bucket policy with a Deny statement for that user or principal will block the action. This is the most likely reason given the scenario.

Answer

The resource ARN for the bucket is incorrect; it should be 'arn:aws:s3:::example-bucket/*'.

Answer

The policy includes s3:GetObject but not s3:ListObjects.

Answer

The policy does not include the s3:ListBucket action.

Question 8

Refer to the exhibit. A data engineer runs a Glue ETL job that reads from a CSV file and writes to a Redshift table. The job fails with the error shown. What is the most likely cause?

Accepted Answer

The source CSV file has fewer columns than the target table.. The error indicates a column count mismatch. The source file has 10 columns but the target expects 12. Option B is wrong because the error is about column count, not data type. Option C is wrong because there is no mention of partition count. Option D is wrong because the error is about validation, not permissions.

Answer

The IAM role for the Glue job does not have permission to write to Redshift.

Answer

The target Redshift table has mismatched data types for some columns.

Answer

The Glue job is using an incorrect number of partitions for the source data.

Question 9

A data engineer is troubleshooting an AWS Glue ETL job that fails with a memory error when processing a large dataset. Which approach can help reduce memory usage?

Accepted Answer

Increase the number of workers for the job. Increasing the number of workers for the job distributes the data processing load across more executors, which reduces the memory burden on each individual worker. In AWS Glue, each worker runs its own Spark executor with a fixed amount of memory; by adding more workers, the data is partitioned across more executors, so each handles a smaller subset of the data. This directly alleviates out-of-memory errors caused by a single worker attempting to process too much data.

Answer

Set the job to use only one worker

Answer

Reduce the number of partitions in the data source

Answer

Increase the worker type to G.2X

Question 10

A company uses Amazon Kinesis Data Streams to ingest real-time clickstream data. The data is consumed by a Lambda function that writes to an S3 bucket. Recently, the Lambda function started timing out. Which step should be taken to resolve this issue?

Accepted Answer

Increase the Lambda function timeout. Increasing the Lambda function timeout is the correct step because the function is timing out, meaning it exceeds the configured maximum execution time. By increasing the timeout, you allow the function more time to process records and write to S3, which can resolve the issue if the processing is simply taking longer than expected. This is a direct fix for timeout errors.

Answer

Set the Lambda reserved concurrency to 1

Answer

Decrease the batch size in the event source mapping

Answer

Increase the number of shards in the Kinesis stream

Question 11

A data engineer is designing a data pipeline that uses AWS Glue to process data from an RDS MySQL database. The pipeline must capture only incremental changes (inserts and updates) and run every hour. Which approach is most cost-effective and reliable?

Accepted Answer

Use Glue job bookmarks to track and process only new and updated records. AWS Glue job bookmarks are a built-in state management feature that persists the last processed state (e.g., JDBC high-water marks or S3 object timestamps) across job runs. For JDBC sources like RDS MySQL, Glue bookmarks track the maximum value of a specified column (typically a primary key or timestamp) and automatically filter subsequent reads to only rows with higher values, capturing inserts and updates without custom code. This is the most cost-effective and reliable approach because it requires no additional infrastructure (unlike DMS), no manual state management (unlike timestamp queries), and no full scans (unlike snapshot comparison).

Answer

Use AWS DMS with change data capture (CDC) to replicate changes to S3

Answer

Add a timestamp column and query rows where timestamp > last run

Answer

Perform a full table scan each hour and compare with previous snapshot

Question 12

A company uses Amazon Kinesis Data Firehose to deliver streaming data to Amazon S3. The delivery stream is failing with 'Insufficient capacity' errors. Which THREE actions should the data engineer take to resolve this issue? (Choose THREE.)

Accepted Answer

Increase the buffer size and buffer interval in the Firehose delivery stream configuration.. Option B is correct because increasing the buffer size and buffer interval in the Firehose delivery stream configuration reduces the frequency of S3 PUT operations, which lowers the throughput demand on the delivery stream and helps alleviate 'Insufficient capacity' errors caused by hitting per-stream or per-account limits. Option D is correct because 'Insufficient capacity' errors often indicate that the account has reached a Kinesis Data Firehose service quota (such as the limit on delivery streams or throughput per stream), so requesting a quota increase via the Service Quotas console directly addresses the root cause. Option E is correct because if the source is a Kinesis data stream, increasing the number of shards raises the stream's ingestion and processing capacity, preventing the upstream stream from being throttled and allowing Firehose to read data without capacity-related failures. Option A is not correct because S3 bucket versioning addresses object overwrite/retention concerns, not Firehose capacity limits, and does not resolve 'Insufficient capacity' errors. Option C is not correct because a CloudWatch alarm only provides monitoring and notification of the error rate; it does not remediate or increase capacity, so it does not resolve the issue.

Answer

Enable S3 bucket versioning to handle concurrent writes.

Answer

Configure a CloudWatch alarm to monitor the error rate.

Question 13

A company has an S3 data lake with millions of objects. A data engineer needs to provide a daily report of objects that are not accessed for 90 days. The engineer must minimize cost and impact on performance. Which approach should be used?

Accepted Answer

Use S3 Storage Lens to generate a dashboard of object age and last access. S3 Storage Lens provides cost-effective analytics including last access date and object age, enabling identification of objects not accessed for 90 days without scanning all objects or incurring high costs. Option A is less suitable because S3 Inventory creates daily lists but requires additional Athena queries, adding complexity and cost. Option B is incorrect because S3 Select is designed for querying object content, not metadata; applying it to millions of objects would be highly inefficient and costly. Option C is incorrect because analyzing S3 server access logs would involve storing and querying large volumes of log data, which is costly and impacts performance, and it logs all requests rather than providing a direct last access timestamp.

Answer

Enable S3 Inventory and query with Athena

Answer

Use S3 Select on each object to check last access metadata

Answer

Analyze S3 server access logs to find objects not accessed

Question 14

A data engineer needs to monitor Amazon DynamoDB table metrics to detect throttled requests. Which CloudWatch metric should the engineer set an alarm on?

Accepted Answer

ThrottledRequests. `ThrottledRequests` is the specific Amazon CloudWatch metric that tracks the number of requests to a DynamoDB table that are throttled due to exceeding the provisioned throughput capacity. This metric directly reflects throttling events, making it the appropriate choice for setting an alarm to detect throttled requests.

Answer

ReadThrottleEvents

Answer

SuccessfulRequestLatency

Answer

ConsumedWriteCapacityUnits

Question 15

A data engineer is designing a data lake on Amazon S3. The data is accessed frequently for the first 30 days, then rarely after that. Which lifecycle policy is MOST cost-effective?

Accepted Answer

Transition to S3 Standard-Infrequent Access (Standard-IA) after 30 days.. Transitioning to S3 Standard-IA after 30 days is the most cost-effective because Standard-IA is designed for data accessed infrequently but requiring rapid access when needed. It has a lower storage cost than S3 Standard but a retrieval fee, making it ideal for data that is frequently accessed for the first 30 days and rarely thereafter. This matches the access pattern exactly.

Answer

Transition to S3 One Zone-IA after 30 days.

Answer

Transition to S3 Glacier Deep Archive after 30 days.

Answer

Keep in S3 Standard for 90 days, then delete.