DEA-C01 Data Store Management Practice Test 17 — 15 Questions

Question 1

A data engineering team is managing an Amazon DynamoDB table that stores user session data. The table has a primary key of user_id (partition key) and session_id (sort key). The application performs strongly consistent reads on individual items. The team notices that read latency increases during peak hours. They suspect that the table is experiencing hot partitions. The team needs to improve read performance without changing the application code. Which solution should they implement?

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

Accepted Answer

Enable DynamoDB Accelerator (DAX) for the table.. DynamoDB Accelerator (DAX) is a fully managed, in-memory cache that delivers up to 10x read performance improvement by caching hot partition data. Since the application uses strongly consistent reads and cannot be changed, DAX provides a drop-in caching layer that reduces read latency on hot partitions without requiring any code modifications.

Answer

Enable DynamoDB global tables to distribute reads across regions.

Answer

Increase the read capacity units for the table.

Answer

Change the application to use eventually consistent reads.

Question 2

A data team uses the CloudFormation template in the exhibit to create an S3 bucket for storing log files. After one year, they notice that the bucket size is larger than expected. They investigate and find that older versions of objects are not being deleted or transitioned. What is the most likely cause?

Accepted Answer

The lifecycle rule does not apply to noncurrent versions because it lacks NoncurrentVersionExpiration or NoncurrentVersionTransition.. The lifecycle rule in the CloudFormation template is missing the `NoncurrentVersionExpiration` or `NoncurrentVersionTransition` properties. When S3 bucket versioning is enabled, lifecycle rules that only specify `ExpirationInDays` or `Transition` apply exclusively to the current version of objects. To manage older (noncurrent) versions, you must explicitly include `NoncurrentVersionExpirationInDays` or `NoncurrentVersionTransitionInDays` in the rule. Without these, noncurrent versions accumulate indefinitely, causing the bucket size to grow larger than expected.

Answer

The lifecycle rule is not enabled because the Status property is not set to 'Enabled' properly.

Answer

The bucket has versioning enabled, but the lifecycle rule only applies to current versions.

Answer

The expiration in days is set to 365, which is too short.

Question 3

A company is migrating its on-premises Oracle data warehouse to Amazon Redshift. The data engineering team needs to load data from Oracle to Redshift using AWS DMS (Database Migration Service). The source database is 2 TB in size. The team wants to minimize downtime and ensure data consistency during full load. Which approach should they take?

Accepted Answer

Use AWS DMS to perform a full load, then enable ongoing replication to capture changes.. AWS DMS can perform a full load of the 2 TB Oracle database while simultaneously capturing ongoing changes via CDC (Change Data Capture). After the full load completes, DMS applies the cached changes to Redshift, ensuring data consistency with minimal downtime. This approach avoids stopping the source database and leverages DMS's native replication capabilities.

Answer

Use UNLOAD command to export data from Oracle to S3, then COPY into Redshift.

Answer

Stop the source database, export data to flat files, upload to S3, and use COPY to load into Redshift.

Answer

Use COPY command directly from Oracle to Redshift over JDBC.

Question 4

A data engineering team is using Amazon EMR to process large datasets stored in Amazon S3. The cluster uses Spot Instances for cost savings. During processing, the team notices that tasks are failing due to Spot Instance interruptions. The team needs to make the EMR job resilient to Spot interruptions without increasing costs significantly. Which solution should they implement?

Accepted Answer

Use EMR instance fleets with a mix of Spot and On-Demand, setting the allocation strategy to 'diversified' and using On-Demand for core nodes.. Using EMR instance fleets with a mix of Spot and On-Demand, setting the allocation strategy to 'diversified', and using On-Demand for core nodes ensures resilience to Spot interruptions without significant cost increase. On-Demand core nodes provide stability for HDFS and critical processing, while diversified allocation for Spot task nodes reduces the risk of simultaneous interruptions. Option A is incorrect because the 'lowest price' allocation strategy prioritizes the cheapest Spot instances, which often have higher interruption rates, and does not protect core nodes. Option B is incorrect because increasing core nodes with On-Demand instances raises costs significantly. Option C is incorrect because using only Spot Instances with automatic termination and checkpointing does not prevent job failures during interruptions; automatic termination would terminate the job, and checkpointing only helps with recovery, not resilience.

Answer

Use EMR instance fleets with a mix of Spot and On-Demand, but set the allocation strategy to 'lowest price'.

Answer

Increase the number of core nodes using On-Demand instances.

Answer

Use only Spot Instances but enable automatic termination and checkpointing.

Question 5

A data engineering team is responsible for an Amazon RDS for PostgreSQL instance that stores financial data. The database is 500 GB in size. The team needs to create a read replica in a different AWS Region for disaster recovery. The source database has automated backups enabled with a retention period of 7 days. The team initiates the cross-region read replica creation. After several hours, the replica status shows 'Replication Lag' of 30 minutes and is increasing. What should the team do to reduce the replication lag?

Accepted Answer

Modify the source DB instance to use a larger instance class.. Modifying the source DB instance to a larger instance class provides more compute and memory resources. If the source instance is resource-constrained, this can reduce I/O contention and allow it to generate and ship write-ahead logs (WAL) more efficiently, potentially decreasing the replication lag to the cross-region read replica. Enabling Multi-AZ on the source database adds a synchronous standby in a different Availability Zone, which does not offload cross-region replication tasks and can increase write latency and resource overhead on the primary, possibly exacerbating replication lag. Deleting and recreating the replica from a snapshot or increasing backup retention does not address ongoing replication performance.

Answer

Delete the replica and create a new one from a snapshot.

Answer

Increase the backup retention period to 35 days.

Answer

Enable Multi-AZ on the source database instance.

Question 6

A data engineering team is using AWS Glue to catalog data in an S3 data lake. They have a Glue crawler that runs daily to update the Data Catalog. Recently, they noticed that the crawler is taking longer to run and sometimes fails because of a timeout. The team suspects the issue is due to the large number of small files in the S3 bucket. They need to improve crawler performance and reliability. Which solution should they implement?

Accepted Answer

Use AWS Glue ETL to consolidate small files into larger ones before crawling.. Consolidating small files into larger ones (e.g., using AWS Glue ETL with a groupFiles or groupSize option, or a separate compaction job) reduces the number of objects the crawler must list and sample. This directly addresses the root cause: a high volume of small files increases metadata operations and can cause crawler timeouts. By reducing file count, the crawler can complete within the default 24-hour timeout and avoid failures.

Answer

Configure the crawler to use a different classifier.

Answer

Increase the crawler timeout to 24 hours.

Answer

Schedule the crawler to run more frequently to avoid large data accumulation.

Question 7

A company uses Amazon Kinesis Data Firehose to deliver streaming data to an Amazon S3 bucket. The data is JSON and each record is about 2 KB. The delivery stream is configured to buffer incoming data to 5 MB or 60 seconds, whichever comes first. The data engineering team notices that the S3 bucket contains many small files (average 2 MB), which makes subsequent processing inefficient. They need to reduce the number of small files without increasing the latency beyond 5 minutes. Which solution should they implement?

Accepted Answer

Increase the buffer size to 50 MB and the buffer interval to 300 seconds.. Increasing the buffer size to 50 MB and the buffer interval to 300 seconds directly addresses the root cause: the current 5 MB buffer size triggers a flush too frequently, producing many 2 MB files. By raising the buffer size to 50 MB, each flush will contain more data, resulting in larger S3 objects (up to ~50 MB uncompressed), while the 300-second interval ensures latency stays within the 5-minute requirement. This reduces the number of small files without requiring additional services or post-processing.

Answer

Enable compression (GZIP) on the delivery stream.

Answer

Use a Lambda function to merge small files after delivery.

Answer

Decrease the buffer size to 1 MB and the buffer interval to 60 seconds.

Question 8

A data engineering team is managing an Amazon DynamoDB table that stores user session data. The table has a primary key of user_id (partition key) and session_id (sort key). The application performs strongly consistent reads on individual items. The team notices that read latency increases during peak hours. They suspect that the table is experiencing hot partitions. The team needs to improve read performance without changing the application code. Which solution should they implement?

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

Accepted Answer

Change the distribution style to EVEN and set the sort key to a date column used in WHERE clauses.. Using EVEN distribution ensures data is evenly distributed across all nodes, avoiding data skew that can occur with KEY distribution on a high-cardinality column like customer_id. Setting the sort key to a date column used in WHERE clauses enables range-restricted scans, significantly reducing the amount of data scanned for common BI queries that filter by date. This combination improves query performance by maximizing parallelism and minimizing I/O.

Answer

Keep the distribution style as AUTO and set the sort key to customer_id.

Answer

Change the distribution style to ALL and set the sort key to customer_id.

Answer

Change the distribution style to KEY on a different column with high cardinality.

Question 9

A data engineering team is using Amazon DynamoDB to store time-series data for a monitoring application. The table has a primary key of device_id (partition key) and timestamp (sort key). The application queries data for a specific device over a time range. The team notices that read latency is high for devices that generate large amounts of data. They need to improve query performance. Which solution should they implement?

Accepted Answer

Create a global secondary index with device_id as partition key and a truncated timestamp as sort key to distribute writes.. Creating a global secondary index (GSI) with device_id as the partition key and a truncated timestamp as the sort key helps distribute write traffic more evenly across partitions. This reduces hot partitions caused by devices that generate large amounts of data, thereby improving read latency by preventing throttling and reducing contention on a single partition.

Answer

Enable DynamoDB Accelerator (DAX) for the table.

Answer

Change the application to use eventually consistent reads.

Answer

Increase the read capacity units for the table.

Question 10

A company runs a MySQL-compatible Amazon Aurora database for its e-commerce platform. The database experiences high write latency during peak hours. The application performs frequent INSERT and UPDATE operations on a table with 50 million rows. The DB instance is db.r5.large with 500 GB of Provisioned IOPS storage. A recent performance analysis shows that the average queue depth is consistently above 32 and write latency exceeds 50 ms. The company needs to reduce write latency without changing the application code. What should a data engineer do?

Accepted Answer

Resize the DB instance to a larger instance type such as db.r5.xlarge.. The high queue depth (consistently above 32) and write latency exceeding 50 ms indicate that the current db.r5.large instance is CPU or I/O constrained for the write workload. Resizing to a larger instance type such as db.r5.xlarge increases the available vCPUs, memory, and network bandwidth, which directly reduces queue depth and write latency by allowing more concurrent write operations to be processed. This solution does not require application code changes and addresses the root cause of insufficient compute capacity for the frequent INSERT and UPDATE operations on the 50-million-row table.

Answer

Enable Aurora Auto Scaling to automatically add reader instances.

Answer

Convert the cluster to Aurora Serverless v2 to automatically scale compute capacity.

Answer

Create a read replica and configure the application to offload read queries.

Question 11

A media company stores video files in an Amazon S3 bucket with S3 Standard storage class. The files are accessed frequently for the first 30 days, then rarely after that. However, the company must be able to restore any deleted file within 7 days. The company wants to minimize storage costs while meeting the access and retention requirements. What should a data engineer do?

Accepted Answer

Use S3 Intelligent-Tiering and enable S3 Versioning on the bucket.. S3 Intelligent-Tiering automatically moves objects between access tiers (frequent, infrequent, and archive instant access) based on changing access patterns, which minimizes storage costs for data with unknown or changing access patterns. Enabling S3 Versioning allows the company to restore any deleted file within 7 days by reverting to a previous version, meeting the retention requirement without additional cost for a separate backup.

Answer

Use S3 Standard-IA storage class from the start.

Answer

Use a lifecycle policy to transition objects to S3 One Zone-IA after 30 days.

Answer

Use S3 Glacier Deep Archive after 30 days and enable S3 Object Lock for retention.

Question 12

A company is migrating an on-premises MySQL database to Amazon RDS for MySQL. The database is 500 GB and has a 24/7 uptime requirement. The migration must minimize downtime. Which approach should be used?

Accepted Answer

Use AWS Database Migration Service (DMS) with ongoing replication to migrate the data.. AWS DMS with ongoing replication (change data capture) allows you to perform a full load of the 500 GB database and then continuously replicate changes from the on-premises MySQL source to the Amazon RDS target. This minimizes downtime because you can cut over to RDS in seconds after the target is synchronized, rather than taking the source offline for an extended period.

Answer

Take a snapshot of the on-premises database, convert it to a volume, and restore to RDS.

Answer

Export the database using mysqldump and import it into RDS using mysql command.

Answer

Create an RDS MySQL read replica from the on-premises database using native replication.

Question 13

A company is using Amazon DynamoDB with on-demand capacity for a gaming application. During a new game launch, write traffic spikes to 50,000 writes per second, but the application experiences throttling. The DynamoDB table has a partition key of 'game_id' and a sort key of 'timestamp'. What is the MOST likely cause of throttling?

Accepted Answer

Hot partitions due to a skewed access pattern on the partition key 'game_id'.. DynamoDB on-demand capacity automatically scales to handle traffic spikes, but it still has per-partition throughput limits. With 'game_id' as the partition key, a single popular game can create a hot partition where all writes target the same partition, exceeding the partition's maximum write capacity (1,000 write capacity units per partition) and causing throttling, even though the overall table capacity is sufficient.

Answer

The table has not enabled auto-scaling for writes.

Answer

The table's on-demand capacity is insufficient for the write spike.

Answer

The sort key is not optimal for write-heavy workloads.

Question 14

A data engineer needs to store semi-structured JSON logs from AWS CloudTrail. The logs are append-only and rarely accessed after 90 days. Which storage solution is MOST cost-effective?

Accepted Answer

Amazon S3 Glacier Deep Archive. Amazon S3 Glacier Deep Archive is the most cost-effective storage solution for CloudTrail logs that are append-only and rarely accessed after 90 days. It offers the lowest storage cost among AWS options (approximately $0.00099 per GB/month) and is designed for data that is accessed at most once or twice per year, with retrieval times of 12–48 hours. Since the logs are rarely accessed after 90 days, the retrieval latency is acceptable, and the cost savings over S3 Standard (which costs ~$0.023 per GB/month) are substantial.

Answer

Amazon S3 Standard

Answer

Amazon EBS with cold HDD volumes

Answer

Amazon DynamoDB with on-demand capacity

Question 15

A company uses Amazon Redshift for data warehousing. The data engineering team notices that queries are slow due to high disk I/O. The team wants to improve query performance without changing the cluster configuration. Which action should the team take?

Accepted Answer

Redesign tables with appropriate sort keys and distribution styles.. Redesigning tables with appropriate sort keys and distribution styles directly addresses high disk I/O by minimizing data scanning and reducing data movement across nodes. Sort keys enable Redshift to skip irrelevant blocks via zone maps, while distribution styles (KEY, ALL, EVEN) optimize data locality for joins and aggregations, reducing I/O without changing cluster configuration.

Answer

Increase the number of nodes in the cluster.

Answer

Run the ANALYZE command to update table statistics.

Answer

Run the VACUUM command to reclaim disk space.