Data-Engineer-Associate Sample Questions Answers

Option A is correct. The MERGE INTO statement is used to conditionally update, insert, or delete rows based on a match condition between a target table and a source table. In this statement, the target table is accounts and the source table is monthly_accounts_update. The join condition is t.customer = s.customer. Because the statement uses WHEN MATCHED THEN DELETE, every row in accounts that has a matching customer value in monthly_accounts_update will be deleted from the target table.

AWS documentation for MERGE INTO states that it conditionally updates, deletes, or inserts rows into an Apache Iceberg table, and the syntax explicitly includes WHEN MATCHED THEN DELETE. AWS Glue guidance and examples for Iceberg also show MERGE INTO as a supported pattern for row-level changes in Glue ETL workflows. This means the syntax is valid for supported Glue-Iceberg use cases, so option D is incorrect. Options B and C are also incorrect because the statement does not retain only matching rows and does not delete the entire table. It deletes only the rows in the target table that satisfy the match condition.

Amazon Athena Federated Query is a feature that allows you to query data from multiple sources using standard SQL. You can use Athena Federated Query to join data from Amazon DynamoDB, Amazon RDS, Amazon Redshift, and Amazon S3, as well as other data sources such as MongoDB, Apache HBase, and Apache Kafka1. Athena Federated Query is a serverless and interactive service, meaning you do not need to provision or manage any infrastructure, and you only pay for the amount of data scanned by your queries. Athena Federated Query is the most cost-effective solution for performing a one-time analysis job on data from multiple sources, as it eliminates the need to copy or move data, and allows you to query data directly from the source.

The other options are not as cost-effective as Athena Federated Query, as they involve additional steps or costs. Option A requires you to provision and pay for an Amazon EMR cluster, which can be expensive and time-consuming for a one-time job. Option B requires you to copy or move data from DynamoDB, RDS, and Redshift to S3, which can incur additional costs for data transfer and storage, and also introduce latency and complexity. Option D requires you to have an existing Redshift cluster, which can be costly and may not be necessary for a one-time job. Option D also does not support querying data from RDS directly, so you would need to use Redshift Federated Query to access RDS data, which adds another layer of complexity2. References:

Amazon Athena Federated Query

Redshift Spectrum vs Federated Query

Option A is the most cost-effective because it enables seamless integration by allowing analysts to access and join external relational data directly from within Amazon Redshift for reporting, without building and operating a separate ingestion pipeline for a one-time (annual) workload. The study material frames Amazon Redshift as the centralized analytics store for complex queries and reporting, making it the natural place to perform the final transformations and analysis.

The alternatives introduce extra operational steps and recurring costs. Option B requires exporting large historical datasets to S3 and then loading them into Redshift, which increases data movement and operational complexity. Option C adds ETL job development, scheduling, retries, and monitoring overhead that is unnecessary if the primary goal is integrated querying for an annual report. Option D (DMS) is best when you need ongoing migration or continuous replication; it is typically more operationally heavy than needed for “query across sources” use cases and still requires additional setup and maintenance.

Because all databases are already in the same VPC and the analytics platform is Redshift, federated querying provides the most direct, lowest-operations path to integrate and analyze the data where it already needs to be consumed.

 A data mesh is an architectural framework that organizes data into domains and treats data as products that are owned and offered for consumption by different teams1. A data mesh requires a centralized layer for data governance and access control, as well as a distributed layer for data storage and analysis. AWS Glue can provide data catalogs and ETL operations for the data mesh, but it cannot provide data governance and access control by itself2. Therefore, the company needs to use another AWS service for this purpose. AWS Lake Formation is a service that allows you to create, secure, and manage data lakes on AWS3. It integrates with AWS Glue and other AWS services to provide centralized data governance and access control for the data mesh. Therefore, option E is correct.

For data storage and analysis, the company can choose from different AWS services depending on their needs and preferences. However, one of the benefits of a data mesh is that it enables data to be stored and processed in a decoupled and scalable way1. Therefore, using serverless or managed services that can handle large volumes and varieties of data is preferable. Amazon S3 is a highly scalable, durable, and secure object storage service that can store any type of data. Amazon Athena is a serverless interactive query service that can analyze data in Amazon S3 using standard SQL. Therefore, option B is a good choice for data storage and analysis in a data mesh. Option A, C, and D are not optimal because they either use relational databases that are not suitable for storing diverse and unstructured data, or they require more management and provisioning than serverless services. References:

1: What is a Data Mesh? - Data Mesh Architecture Explained - AWS

2: AWS Glue - Developer Guide

3: AWS Lake Formation - Features

[4]: Design a data mesh architecture using AWS Lake Formation and AWS Glue

[5]: Amazon S3 - Features

[6]: Amazon Athena - Features

The most efficient and low-operational-overhead solution for ingesting data into Amazon Redshift from Amazon Kinesis Data Streams is to use Amazon Redshift streaming ingestion. This feature allows Redshift to directly ingest streaming data from Kinesis Data Streams and process it in real-time.

Amazon Redshift Streaming Ingestion:

Redshift supports native streaming ingestion from Kinesis Data Streams, allowing real-time data to be queried using materialized views.

This solution reduces operational complexity because you don ' t need intermediary services like Amazon Kinesis Data Firehose or S3 for batch loading.

Apache Iceberg is a table format designed for large-scale data lakes that supports ACID transactions, schema evolution, time travel, and row-level updates and deletes. Using S3 Tables with Apache Iceberg provides a fully managed experience that integrates natively with Amazon Athena, Amazon Redshift, and Amazon EMR.

By using AWS Glue with the Iceberg catalog, the data engineer can perform daily updates and deletions without managing Spark clusters, compaction scheduling, or metadata cleanup manually. Iceberg handles snapshots, file pruning, and unreferenced file removal automatically, significantly reducing operational overhead.

Apache Hudi requires Amazon EMR clusters, Spark jobs, and manual compaction orchestration, increasing complexity. The Parquet-only approaches in options C and D do not support updates or deletes efficiently and would require full rewrites of datasets, which is not scalable.

Therefore, using S3 Tables with Apache Iceberg provides the most efficient, scalable, and low-maintenance solution that satisfies all query and update requirements.

Athena workgroups are a way to isolate query execution and query history among users, teams, and applications that share the same AWS account. By creating a workgroup for each use case, the company can control the access and actions on the workgroup resource using resource-level IAM permissions or identity-based IAM policies. The company can also use tags to organize and identify the workgroups, and use them as conditions in the IAM policies to grant or deny permissions to the workgroup. This solution meets the requirements of separating query processes and access to query history among users, teams, and applications that are in the same AWS account. References:

Athena Workgroups

IAM policies for accessing workgroups

Workgroup example policies

Option D is correct because Aurora zero-ETL integration with Amazon Redshift is the AWS-managed feature built specifically to make transactional Aurora data available in Amazon Redshift Serverless in near real time. AWS documents that Aurora zero-ETL is a fully managed solution that makes transactional data available in the analytics destination after it is written to the Aurora cluster, eliminating the need to build and maintain complex ETL pipelines. AWS also states that Redshift zero-ETL supports a target data warehouse that can be a Redshift Serverless workgroup, which directly matches the question.

Option A and C add more operational components, such as DMS tasks, replication design, schema handling, and additional streaming infrastructure. Option B is also more operationally heavy because MSK Connect with Debezium requires Kafka-based CDC infrastructure and connector management. Since the requirement emphasizes serverless, near real-time analytics, and least operational overhead, the native zero-ETL integration is the best fit. This also aligns with the study guide’s focus on choosing AWS services that minimize management effort while supporting ingestion and analytics workflows.

Amazon Athena charges per amount of data scanned by each query. When the same query with identical parameters is executed multiple times, re-scanning the data repeatedly increases both cost and execution time. Athena provides query result reuse, which allows previously computed query results to be reused when an identical query is rerun within a configurable time window.

By enabling query result reuse at the Athena workgroup level, Athena automatically returns cached results instead of re-reading data from Amazon S3. This significantly improves query performance while reducing data scan costs. This feature is specifically designed for workloads where analysts frequently rerun complex queries with the same parameters, which directly matches the scenario.

Converting data to JSON would degrade performance because JSON is not a columnar format and increases scan size. Using Amazon EMR introduces unnecessary infrastructure management and cost for a query optimization problem that Athena already solves natively. Amazon Redshift Spectrum is intended for querying S3 data from within Redshift and does not address repeated-query optimization in Athena.

Therefore, configuring query result reuse in the Athena workgroup is the most cost-effective and operationally efficient solution.

The NEST TO MAP transformation allows you to combine multiple columns into a single column that contains a JSON object with key-value pairs. This is the easiest way to achieve the desired format for the physical address data, as you can simply select the columns to nest and specify the keys for each column. The NEST TO ARRAY transformation creates a single column that contains an array of values, which is not the same as the JSON object format. The PIVOT transformation reshapes the data by creating new columns from unique values in a selected column, which is not applicable for this use case. Writing a Lambda function in Python requires more coding effort than using AWS Glue DataBrew, which provides a visual and interactive interface for data transformations. References:

7 most common data preparation transformations in AWS Glue DataBrew (Section: Nesting and unnesting columns)

NEST TO MAP - AWS Glue DataBrew (Section: Syntax)

To build an enterprise data catalog with metadata for storage formats, the easiest and most efficient solution is using an AWS Glue crawler. The Glue crawler can scan Amazon S3 buckets and Amazon RDS databases to automatically create a data catalog that includes metadata such as the schema and storage format (e.g., CSV, Parquet, etc.). By using AWS Glue crawler classifiers, you can configure the crawler to recognize the format of the data and store this information directly in the catalog.

Option B: Use an AWS Glue crawler to build a data catalog. Use AWS Glue crawler classifiers to recognize the format of data and store the format in the catalog.This option meets the requirements with the least effort because Glue crawlers automate the discovery and cataloging of data from multiple sources, including S3 and RDS, while recognizing various file formats via classifiers.

Other options (A, C, D) involve additional manual steps, like having data stewards inspect the data, or using services like Amazon Macie that focus more on sensitive data detection rather than format cataloging.

Option B is correct because IAM policies are the standard AWS mechanism to control access to Amazon S3, and SSE-KMS encrypts S3 data with AWS KMS keys. AWS S3 documentation for SSE-KMS explains that it uses AWS Key Management Service keys for server-side encryption, and Secrets Manager documentation states that you can store secrets securely and configure automatic rotation. AWS further explains that rotation updates the credentials in both the secret and the target database or service, and that Secrets Manager uses a Lambda rotation function for supported rotation patterns.

Option A is weaker because Lambda environment variables are not the right service for managed secret storage and automatic rotation. Option C is not the best answer because S3 ACLs are not the preferred modern access-control model compared with IAM and bucket policies, and Parameter Store is not the main AWS service for built-in managed database credential rotation. Option D uses SSE-S3, not KMS-based encryption, and relies on a custom rotation approach instead of the native managed secret-rotation service. The study guide also highlights AWS KMS for encryption-key management and Secrets Manager for rotating credentials.

The best solution to meet the requirements of creating a data catalog that includes the IoT data, and allowing the analytics department to index the data, most cost-effectively, is to create an Amazon Athena workgroup, explore the data that is in Amazon S3 by using Apache Spark through Athena, and provide the Athena workgroup schema and tables to the analytics department.

Amazon Athena is a serverless, interactive query service that makes it easy to analyze data directly in Amazon S3 using standard SQL or Python1. Amazon Athena also supports Apache Spark, an open-source distributed processing framework that can run large-scale data analytics applications across clusters of servers2. You can use Athena to run Spark code on data in Amazon S3 without having to set up, manage, or scale any infrastructure. You can also use Athena to create and manage external tables that point to your data in Amazon S3, and store them in an external data catalog, such as AWS Glue Data Catalog, Amazon Athena Data Catalog, or your own Apache Hive metastore3. You can create Athena workgroups to separate query execution and resource allocation based on different criteria, such as users, teams, or applications4. You can share the schemas and tables in your Athena workgroup with other users or applications, such as Amazon QuickSight, for data visualization and analysis5.

Using Athena and Spark to create a data catalog and explore the IoT data in Amazon S3 is the most cost-effective solution, as you pay only for the queries you run or the compute you use, and you pay nothing when the service is idle1. You also save on the operational overhead and complexity of managing data warehouse infrastructure, as Athena and Spark are serverless and scalable. You can also benefit from the flexibility and performance of Athena and Spark, as they support various data formats, including JSON, and can handle schema changes and complex queries efficiently.

Option A is not the best solution, as creating an AWS Glue Data Catalog, configuring an AWS Glue Schema Registry, creating a new AWS Glue workload to orchestrate the ingestion of the data that the analytics department will use into Amazon Redshift Serverless, would incur more costs and complexity than using Athena and Spark. AWS Glue Data Catalog is a persistent metadata store that contains table definitions, job definitions, and other control information to help you manage your AWS Glue components6. AWS Glue Schema Registry is a service that allows you to centrally store and manage the schemas of your streaming data in AWS Glue Data Catalog7. AWS Glue is a serverless data integration service that makes it easy to prepare, clean, enrich, and move data between data stores8. Amazon Redshift Serverless is a feature of Amazon Redshift, a fully managed data warehouse service, that allows you to run and scale analytics without having to manage data warehouse infrastructure9. While these services are powerful and useful for many data engineering scenarios, they are not necessary or cost-effective for creating a data catalog and indexing the IoT data in Amazon S3. AWS Glue Data Catalog and Schema Registry charge you based on the number of objects stored and the number of requests made67. AWS Glue charges you based on the compute time and the data processed by your ETL jobs8. Amazon Redshift Serverless charges you based on the amount of data scanned by your queries and the compute time used by your workloads9. These costs can add up quickly, especially if you have large volumes of IoT data and frequent schema changes. Moreover, using AWS Glue and Amazon Redshift Serverless would introduce additional latency and complexity, as you would have to ingest the data from Amazon S3 to Amazon Redshift Serverless, and then query it from there, instead of querying it directly from Amazon S3 using Athena and Spark.

Option B is not the best solution, as creating an Amazon Redshift provisioned cluster, creating an Amazon Redshift Spectrum database for the analytics department to explore the data that is in Amazon S3, and creating Redshift stored procedures to load the data into Amazon Redshift, would incur more costs and complexity than using Athena and Spark. Amazon Redshift provisioned clusters are clusters that you create and manage by specifying the number and type of nodes, and the amount of storage and compute capacity10. Amazon Redshift Spectrum is a feature of Amazon Redshift that allows you to query and join data across your data warehouse and your data lake using standard SQL11. Redshift stored procedures are SQL statements that you can define and store in Amazon Redshift, and then call them by using the CALL command12. While these features are powerful and useful for many data warehousing scenarios, they are not necessary or cost-effective for creating a data catalog and indexing the IoT data in Amazon S3. Amazon Redshift provisioned clusters charge you based on the node type, the number of nodes, and the duration of the cluster10. Amazon Redshift Spectrum charges you based on the amount of data scanned by your queries11. These costs can add up quickly, especially if you have large volumes of IoT data and frequent schema changes. Moreover, using Amazon Redshift provisioned clusters and Spectrum would introduce additional latency and complexity, as you would have to provision and manage the cluster, create an external schema and database for the data in Amazon S3, and load the data into the cluster using stored procedures, instead of querying it directly from Amazon S3 using Athena and Spark.

Option D is not the best solution, as creating an AWS Glue Data Catalog, configuring an AWS Glue Schema Registry, creating AWS Lambda user defined functions (UDFs) by using the Amazon Redshift Data API, and creating an AWS Step Functions job to orchestrate the ingestion of the data that the analytics department will use into Amazon Redshift Serverless, would incur more costs and complexity than using Athena and Spark. AWS Lambda is a serverless compute service that lets you run code without provisioning or managing servers13. AWS Lambda UDFs are Lambda functions that you can invoke from within an Amazon Redshift query. Amazon Redshift Data API is a service that allows you to run SQL statements on Amazon Redshift clusters using HTTP requests, without needing a persistent connection. AWS Step Functions is a service that lets you coordinate multiple AWS services into serverless workflows. While these services are powerful and useful for many data engineering scenarios, they are not necessary or cost-effective for creating a data catalog and indexing the IoT data in Amazon S3. AWS Glue Data Catalog and Schema Registry charge you based on the number of objects stored and the number of requests made67. AWS Lambda charges you based on the number of requests and the duration of your functions13. Amazon Redshift Serverless charges you based on the amount of data scanned by your queries and the compute time used by your workloads9. AWS Step Functions charges you based on the number of state transitions in your workflows. These costs can add up quickly, especially if you have large volumes of IoT data and frequent schema changes. Moreover, using AWS Glue, AWS Lambda, Amazon Redshift Data API, and AWS Step Functions would introduce additional latency and complexity, as you would have to create and invoke Lambda functions to ingest the data from Amazon S3 to Amazon Redshift Serverless using the Data API, and coordinate the ingestion process using Step Functions, instead of querying it directly from Amazon S3 using Athena and Spark. References:

What is Amazon Athena?

Apache Spark on Amazon Athena

Creating tables, updating the schema, and adding new partitions in the Data Catalog from AWS Glue ETL jobs

Managing Athena workgroups

Using Amazon QuickSight to visualize data in Amazon Athena

AWS Glue Data Catalog

AWS Glue Schema Registry

What is AWS Glue?

Amazon Redshift Serverless

Amazon Redshift provisioned clusters

Querying external data using Amazon Redshift Spectrum

Using stored procedures in Amazon Redshift

What is AWS Lambda?

[Creating and using AWS Lambda UDFs]

[Using the Amazon Redshift Data API]

[What is AWS Step Functions?]

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide

This solution will meet the requirements with the least operational overhead because it uses the AWS Glue Data Catalog as the central metadata repository for data sources that run in the AWS Cloud. The AWS Glue Data Catalog is a fully managed service that provides a unified view of your data assets across AWS and on-premises data sources. It stores the metadata of your data in tables, partitions, and columns, and enables you to access and query your data using various AWS services, such as Amazon Athena, Amazon EMR, and Amazon Redshift Spectrum. You can use AWS Glue crawlers to connect to multiple data stores, such as Amazon RDS, Amazon Redshift, and Amazon S3, and to update the Data Catalog with metadata changes. AWS Glue crawlers can automatically discover the schema and partition structure of your data, and create or update the corresponding tables in the Data Catalog. You can schedule the crawlers to run periodically to update the metadata catalog, and configure them to detect changes to the source metadata, such as new columns, tables, or partitions12.

The other options are not optimal for the following reasons:

A. Use Amazon Aurora as the data catalog. Create AWS Lambda functions that will connect to the data catalog. Configure the Lambda functions to gather the metadata information from multiple sources and to update the Aurora data catalog. Schedule the Lambda functions to run periodically. This option is not recommended, as it would require more operational overhead to create and manage an Amazon Aurora database as the data catalog, and to write and maintain AWS Lambda functions to gather and update the metadata information from multiple sources. Moreover, this option would not leverage the benefits of the AWS Glue Data Catalog, such as data cataloging, data transformation, and data governance.

C. Use Amazon DynamoDB as the data catalog. Create AWS Lambda functions that will connect to the data catalog. Configure the Lambda functions to gather the metadata information from multiple sources and to update the DynamoDB data catalog. Schedule the Lambda functions to run periodically. This option is also not recommended, as it would require more operational overhead to create and manage an Amazon DynamoDB table as the data catalog, and to write and maintain AWS Lambda functions to gather and update the metadata information from multiple sources. Moreover, this option would not leverage the benefits of the AWS Glue Data Catalog, such as data cataloging, data transformation, and data governance.

D. Use the AWS Glue Data Catalog as the central metadata repository. Extract the schema for Amazon RDS and Amazon Redshift sources, and build the Data Catalog. Use AWS Glue crawlers for data that is in Amazon S3 to infer the schema and to automatically update the Data Catalog. This option is not optimal, as it would require more manual effort to extract the schema for Amazon RDS and Amazon Redshift sources, and to build the Data Catalog. This option would not take advantage of the AWS Glue crawlers’ ability to automatically discover the schema and partition structure of your data from various data sources, and to create or update the corresponding tables in the Data Catalog.

1: AWS Glue Data Catalog

2: AWS Glue Crawlers

Amazon Aurora

AWS Lambda

Amazon DynamoDB

To create a new table named cities_usa in Amazon Athena based on a subset of data from the existing cities_world table, you should use an INSERT INTO statement combined with a SELECT statement to filter only the records where the country is ' usa ' . The correct SQL syntax would be:

Option A: INSERT INTO cities_usa (city, state) SELECT city, state FROM cities_world WHERE country= ' usa ' ;This statement inserts only the cities and states where the country column has a value of ' usa ' from the cities_world table into the cities_usa table. This is a correct approach to create a new table with data filtered from an existing table in Athena.

Options B, C, and D are incorrect due to syntax errors or incorrect SQL usage (e.g., the MOVE command or the use of UPDATE in a non-relevant context).

AWS Lake Formation is specifically designed to simplify fine-grained access control for data lakes stored in Amazon S3. By using Lake Formation tag-based access control (LF-TBAC), administrators can define access policies once and apply them dynamically based on tags rather than managing individual IAM policies.

LF-Tags can be assigned to databases, tables, and columns in the AWS Glue Data Catalog. Departments are granted permissions based on tags, ensuring that each department can access only the data it is authorized to view. This approach scales efficiently as datasets and departments grow, which is critical for data lakes ingesting millions of records daily.

Managing IAM policies per department is operationally complex and error-prone. Redshift-based access centralization limits flexibility and introduces unnecessary infrastructure. Syncing data into an RDS database adds cost, latency, and maintenance overhead.

Lake Formation provides centralized governance, auditing, and least-privilege enforcement with minimal administrative effort, making it the optimal solution.

The company ' s Apache Spark ETL job on Amazon EMR uses high CPU but low memory, meaning that compute-optimized EC2 instances would be the most cost-effective choice. These instances are designed for high-performance compute applications, where CPU usage is high, but memory needs are minimal, which is exactly the case here.

Compute Optimized Instances:

Compute-optimized instances, such as the C5 series, provide a higher ratio of CPU to memory, which is more suitable for jobs with high CPU usage and relatively low memory consumption.

Switching from general-purpose EC2 instances to compute-optimized instances can reduce costs while improving performance, as these instances are optimized for workloads like Spark jobs that perform a lot of computation.

Step 1: Understanding the Problem

The company collects clickstream data via Amazon Kinesis Data Streams and stores it in JSON format in Amazon S3 using Kinesis Data Firehose. They use Amazon Athena to query the data, but they want to reduce Athena costs while maintaining the same data pipeline.

Since Athena charges based on the amount of data scanned during queries, reducing the data size (by converting JSON to a more efficient format like Apache Parquet) is a key solution to lowering costs.

Step 2: Why Option A is Correct

Option A provides a straightforward way to reduce costs with minimal management overhead:

Changing the Firehose output format to Parquet: Parquet is a columnar data format, which is more compact and efficient than JSON for Athena queries. It significantly reduces the amount of data scanned, which in turn reduces Athena query costs.

Custom S3 Object Prefix (YYYYMMDD): Adding a date-based prefix helps in partitioning the data, which further improves query efficiency in Athena by limiting the data scanned to only relevant partitions.

AWS Glue ETL Job for Existing Data: To handle existing data stored in JSON format, a one-time AWS Glue ETL job can combine small JSON files, convert them to Parquet, and apply the YYYYMMDD prefix. This ensures consistency in the S3 bucket structure and allows Athena to efficiently query historical data.

ALTER TABLE ADD PARTITION: This command updates Athena ' s table metadata to reflect the new partitions, ensuring that future queries target only the required data.

Step 3: Why Other Options Are Not Ideal

Option B (Apache Spark on EMR) introduces higher management effort by requiring the setup of Apache Spark jobs and an Amazon EMR cluster. While it achieves the goal of converting JSON to Parquet, it involves running and maintaining an EMR cluster, which adds operational complexity.

Option C (Kinesis and Apache Flink) is a more complex solution involving Apache Flink, which adds a real-time streaming layer to aggregate data. Although Flink is a powerful tool for stream processing, it adds unnecessary overhead in this scenario since the company already uses Kinesis Data Firehose for batch delivery to S3.

Option D (AWS Lambda with Firehose) suggests using AWS Lambda to convert records in real time. While Lambda can work in some cases, it ' s generally not the best tool for handling large-scale data transformations like JSON-to-Parquet conversion due to potential scaling and invocation limitations. Additionally, running parallel Glue jobs further complicates the setup.

Step 4: How Option A Minimizes Costs

By using Apache Parquet, Athena queries become more efficient, as Athena will scan significantly less data, directly reducing query costs.

Firehose natively supports Parquet as an output format, so enabling this conversion in Firehose requires minimal effort. Once set, new data will automatically be stored in Parquet format in S3, without requiring any custom coding or ongoing management.

The AWS Glue ETL job for historical data ensures that existing JSON files are also converted to Parquet format, ensuring consistency across the data stored in S3.

Conclusion:

Option A meets the requirement to reduce Athena costs without recreating the data pipeline, using Firehose’s native support for Apache Parquet and a simple one-time AWS Glue ETL job for existing data. This approach involves minimal management effort compared to the other solutions.

Option A is correct because when many AWS Glue jobs start at the same time in the same subnet, Glue needs to create network interfaces in that subnet. If the subnet does not have enough free private IP addresses available, jobs can fail with exactly this type of error. AWS service documentation for VPC-based managed data services consistently notes that if there is no available free IP address in a specified subnet, the service cannot create or add the required ENIs and the workload can fail or degrade.

This is a subnet-level exhaustion issue, not a VPC-wide issue. A VPC can still have free addresses in other subnets while the specific subnet chosen for the jobs is out of usable addresses. That makes D incorrect. Option B is incorrect because G.1X is a valid Glue worker type and does not inherently prevent subnet access. Option C is also incorrect because the error is about address availability, not IAM authorization. In practice, this problem is usually resolved by using a larger subnet or spreading workloads across additional subnets so enough IPs are available when multiple Glue jobs launch concurrently. This aligns with the exam’s focus on networking capacity planning for managed data services.

The correct answer is to configure AWS Glue triggers to run the ETL jobs every hour and use AWS Glue connections to establish connectivity between the data sources and Amazon Redshift. AWS Glue triggers are a way to schedule and orchestrate ETL jobs with the least operational overhead. AWS Glue connections are a way to securely connect to data sources and targets using JDBC or MongoDB drivers. AWS Glue DataBrew is a visual data preparation tool that does not support MongoDB as a data source. AWS Lambda functions are a serverless option to schedule and run ETL jobs, but they have a limit of 15 minutes for execution time, which may not be enough for complex transformations. The Redshift Data API is a way to run SQL commands on Amazon Redshift clusters without needing a persistent connection, but it does not support loading data from AWS Glue ETL jobs. References:

AWS Glue triggers

AWS Glue connections

AWS Glue DataBrew

[AWS Lambda functions]

[Redshift Data API]

Option C is correct because the company only needs to keep the original files for 90 days and then delete them. AWS states that S3 Lifecycle can be used to delete expired objects automatically, which directly matches this requirement. There is no need to transition the objects to another storage class because the files are not needed beyond day 90. Adding a transition step would add complexity and, in this case, would not improve cost-effectiveness.

Option A is not optimal because S3 Glacier Flexible Retrieval has a 90-day minimum storage duration and is intended for longer-term archival. Transitioning objects right before deleting them is unnecessary overhead and can add minimum-duration charges. Option B is incorrect because Object Lock is for retention and write-once-read-many protection, not for normal temporary retention with lowest cost. Option D is also less cost-effective because S3 Intelligent-Tiering is useful when access patterns are uncertain, but here the retention period and deletion timing are already known. The simplest and cheapest solution is to keep the originals in S3 Standard for the required 90 days and then expire them with Lifecycle.

The original query does not return results for all of the products because the year column in the sales_data table is not an integer, but a timestamp. Therefore, the WHERE clause does not filter the data correctly, and only returns the products that have a null value for the year column. To fix this, the data engineer should use the extract function to extract the year from the timestamp and compare it with 2023. This way, the query will return the correct results for all of the products in the sales_data table. The other options are either incorrect or irrelevant, as they do not address the root cause of the issue. Replacing sum with count does not change the filtering condition, adding HAVING clause does not affect the grouping logic, and removing the GROUP BY clause does not solve the problem of missing products. References:

Troubleshooting JSON queries - Amazon Athena (Section: JSON related errors)

When I query a table in Amazon Athena, the TIMESTAMP result is empty (Section: Resolution)

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide (Chapter 7, page 197)

The Amazon Redshift Data API enables you to interact with your Amazon Redshift data warehouse in an easy and secure way. You can use the Data API to run SQL commands, such as loading data into tables, without requiring a persistent connection to the cluster. The Data API also integrates with Amazon EventBridge, which allows you to monitor the execution status of your SQL commands and trigger actions based on events. By using the Data API to publish an event to EventBridge, the data engineer can invoke the Lambda function that writes the load statuses to the DynamoDB table. This solution is scalable, reliable, and cost-effective. The other options are either not possible or not optimal. You cannot use a second Lambda function to invoke the first Lambda function based on CloudWatch or CloudTrail events, as these services do not capture the load status of Redshift tables. You can use the Data API to publish a message to an SQS queue, but this would require additional configuration and polling logic to invoke the Lambda function from the queue. This would also introduce additional latency and cost. References:

Using the Amazon Redshift Data API

Using Amazon EventBridge with Amazon Redshift

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide, Chapter 2: Data Store Management, Section 2.2: Amazon Redshift

For near real-time processing of new files in S3, event-driven ingestion is optimal. S3 Event Notifications can trigger AWS Lambda to immediately load data into Amazon Redshift, eliminating latency associated with batch scheduling.

“Event-based triggers using S3 notifications and Lambda functions are effective for near real-time ingestion pipelines into Amazon Redshift.”

– Ace the AWS Certified Data Engineer - Associate Certification - version 2 - apple.pdf

Option C (Glue bookmarks) is best for batch jobs, and zero-ETL applies to Aurora to Redshift, not S3.

AWS Glue DataBrew provides a no-code way to define and run data quality rulesets for data stored in S3. You can trigger profiling jobs via Amazon EventBridge on new uploads for automated checks.

“Use AWS Glue DataBrew to define and run data quality rules on S3 datasets with minimal coding effort. Automate validation by triggering jobs through EventBridge.”

– Ace the AWS Certified Data Engineer - Associate Certification - version 2 - apple.pdf

To meet both encryption at rest and in transit, a single Amazon EMR security configuration can be created specifying the AWS KMS key for encryption at rest and the PEM file for in-transit encryption. The study guide clearly states:

“AWS Key Management Service (KMS) provides encryption for data at rest, and SSL/TLS ensures encryption for data in transit, providing end-to-end encryption within an AWS environment.”

– Ace the AWS Certified Data Engineer - Associate Certification - version 2 - apple.pdf

A single security configuration is sufficient and the cleanest way to apply these security features during EMR cluster setup.

The ecommerce company wants to migrate its data pipelines into the AWS Cloud without managing servers, and the solution must orchestrate Python and Bash scripts without refactoring code. Amazon Managed Workflows for Apache Airflow (Amazon MWAA) is the most suitable solution for this scenario.

Option B: Amazon Managed Workflows for Apache Airflow (Amazon MWAA)MWAA is a managed orchestration service that supports Python and Bash scripts via Directed Acyclic Graphs (DAGs) for workflows. It is a serverless, managed version of Apache Airflow, which is commonly used for orchestrating complex data workflows, making it an ideal choice for migrating existing pipelines without refactoring. It supports Python, Bash, and other scripting languages, and the company would not need to manage the underlying infrastructure.

Other options:

AWS Lambda (Option A) is more suited for event-driven workflows but would require breaking down the pipeline into individual Lambda functions, which may require refactoring.

AWS Step Functions (Option C) is good for orchestration but lacks native support for Python and Bash without using Lambda functions, and it may require code changes.

AWS Glue (Option D) is an ETL service primarily for data transformation and not suitable for orchestrating general scripts without modification.

In Amazon Redshift, a compound sort key is designed to optimize the performance of queries that use filtering and join conditions on the columns in the sort key. A compound sort key orders the data based on the first column, followed by the second, and so on. In the scenario given, the compound sort key consists of Region ID, Department ID, and Role ID. Therefore, queries that filter on the leading columns of the sort key are more likely to benefit from this order.

Option B: " Select * from Employee where Region ID= ' North America ' and Department ID=20; " This query will perform well because it uses both the Region ID and Department ID, which are the first two columns of the compound sort key. The order of the columns in the WHERE clause matches the order in the sort key, thus allowing the query to scan fewer rows and improve performance.

Option C: " Select * from Employee where Department ID=20 and Region ID= ' North America ' ; " This query also benefits from the compound sort key because it includes both Region ID and Department ID, which are the first two columns in the sort key. Although the order in the WHERE clause does not match exactly, Amazon Redshift will still leverage the sort key to reduce the amount of data scanned, improving query speed.

Options A, D, and E are less optimal because they do not utilize the sort key as effectively:

Option A only filters by the Region ID, which may still use the sort key but does not take full advantage of the compound nature.

Option D uses only Role ID, the last column in the compound sort key, which will not benefit much from sorting since it is the third key in the sort order.

Option E filters on Region ID and Role ID but skips the Department ID column, making it less efficient for the compound sort key.

Option B is correct because access to specific S3 folders for AWS Glue jobs should be restricted by scoping permissions to the correct S3 object ARNs in the IAM policy Resource element, such as bucket paths with wildcard patterns for the allowed prefixes. Time-based access is then enforced by adding a Condition that uses the global IAM condition key aws:CurrentTime with date operators such as DateGreaterThan and DateLessThan. AWS IAM documentation explicitly states that date condition operators are used with aws:CurrentTime or aws:EpochTime in the Condition element.

Option C is not the best answer because s3:prefix is generally used with bucket listing operations rather than as the main mechanism to control object-level access for Glue reads and writes. For object access, the cleanest least-privilege design is to scope the allowed object ARNs directly. Option A is incorrect because $util.time is not an IAM policy condition key. Option D uses the wrong S3 condition key for the stated requirement. The uploaded study guide also emphasizes granting AWS Glue access through an IAM service role with the right scoped permissions, which fits this pattern.

Using an Amazon EventBridge rule to run an AWS Glue job or invoke an AWS Glue workflow job every 15 minutes are two possible solutions that will meet the requirements. AWS Glue is a serverless ETL service that can process and load data from various sources to various targets, including Amazon Redshift. AWS Glue can handle different data formats, such as CSV, JSON, and Parquet, and also support schema evolution, meaning it can adapt to changes in the data schema over time. AWS Glue can also leverage Apache Spark to perform distributed processing and transformation of large datasets. AWS Glue integrates with Amazon EventBridge, which is a serverless event bus service that can trigger actions based on rules and schedules. By using an Amazon EventBridge rule, you can invoke an AWS Glue job or workflow every 15 minutes, and configure the job or workflow to run an AWS Glue crawler and then load the data into the Amazon Redshift tables. This way, you can build a cost-effective and scalable ETL pipeline that can handle data from 10 source systems and function correctly despite changes to the data schema.

The other options are not solutions that will meet the requirements. Option C, configuring an AWS Lambda function to invoke an AWS Glue crawler when a file is loaded into the S3 bucket, and creating a second Lambda function to run the AWS Glue job, is not a feasible solution, as it would require a lot of Lambda invocations and coordination. AWS Lambda has some limits on the execution time, memory, and concurrency, which can affect the performance and reliability of the ETL pipeline. Option D, configuring an AWS Lambda function to invoke an AWS Glue workflow when a file is loaded into the S3 bucket, is not a necessary solution, as you can use an Amazon EventBridge rule to invoke the AWS Glue workflow directly, without the need for a Lambda function. Option E, configuring an AWS Lambda function to invoke an AWS Glue job when a file is loaded into the S3 bucket, and configuring the AWS Glue job to put smaller partitions of the DataFrame into an Amazon Kinesis Data Firehose delivery stream, is not a cost-effective solution, as it would incur additional costs for Lambda invocations and data delivery. Moreover, using Amazon Kinesis Data Firehose to load data into Amazon Redshift is not suitable for frequent and small batches of data, as it can cause performance issues and data fragmentation. References:

AWS Glue

Amazon EventBridge

Using AWS Glue to run ETL jobs against non-native JDBC data sources

[AWS Lambda quotas]

[Amazon Kinesis Data Firehose quotas]

Option A is the best answer because AWS DMS documentation recommends that, when practical, you create the replication instance in the same Region as your target endpoint, and in the same VPC or subnet as your target endpoint. In this scenario, the target is the Amazon Redshift cluster in eu-west-1 in Account_B, so placing the replication instance there is the most appropriate design. AWS DMS uses the replication instance to connect to the source, read the source data, transform it as needed, and load it into the target.

This also makes architectural sense because the replication instance must have network connectivity to both endpoints, and colocating it with the Redshift target usually simplifies connectivity to the target warehouse and aligns with AWS guidance. A replication instance in Account_B, eu-west-1 can still connect to the PostgreSQL source in Account_A, eu-east-1, provided the required networking and permissions are configured. AWS DMS supports replication tasks by defining a replication instance plus source and target endpoints.

The study guide also identifies AWS DMS as the correct managed service for database migration and continuous replication scenarios, including full load and change data capture. That matches this cross-account, cross-Region migration use case.

Option A is the most direct and operationally efficient way to trigger the existing Lambda-based load into Amazon Redshift whenever a new CSV file is uploaded to Amazon S3. The key requirement is event-driven automation on “new object created” events, and the study material explicitly describes using Amazon EventBridge to react to S3 uploads by creating a rule on the default event bus and routing matching events to a target application.

Compared with queue-based designs (Options B and D), Option A reduces components: there is no need to manage an SQS queue, batching/visibility timeout behavior, or retry semantics at the consumer layer. This is especially appropriate because the files are small (≤ 5 MB) and the Lambda function is already implemented to perform the copy/load step. Using DMS (Option C) is not designed for “S3 object arrival → Lambda → Redshift” event triggering; it introduces unnecessary services and operational work for a simple file-ingestion trigger.

Therefore, integrating S3 event notifications with EventBridge and targeting the Lambda function best meets the requirement with the least added complexity.

Step 1: Log Collection (FluentBit and CloudWatch)

Option A suggests using FluentBit to collect logs and OpenTelemetry to collect traces.

FluentBit is a lightweight log processor that integrates with Amazon EKS to collect and forward logs from Kubernetes clusters. It is widely used with minimal overhead, making it an ideal choice for log collection in this scenario. FluentBit is also natively compatible with AWS services.

OpenTelemetry is a popular framework to collect traces from distributed applications. It provides observability, making it easier to monitor microservices.

This combination allows you to effectively gather both logs and traces with minimal setup and configuration, aligning with the goal of least development effort.

CloudWatch can be used to monitor logs (Option B and C). However, for applications that need more custom and fine-grained control over logging mechanisms, FluentBit and OpenTelemetry are the preferred choice in microservice environments.

Step 2: Log and Trace Correlation (Amazon OpenSearch)

Option D (Amazon OpenSearch) is specifically designed to search, analyze, and visualize logs, metrics, and traces in real-time. OpenSearch allows you to correlate logs and traces effectively.

With Amazon OpenSearch, you can set up dashboards that help in visualizing both logs and traces together, which assists in identifying any failure points across the entire request flow.

It offers integrations with FluentBit and OpenTelemetry, ensuring that both logs from the EKS cluster and application traces are centrally collected, stored, and correlated without additional heavy development.

Step 3: Why Other Options Are Not Suitable

Option B (Amazon Kinesis) is designed for real-time data streaming and analytics but is not as well-suited for tracing microservice requests and logs correlation compared to OpenSearch.

Option C (Amazon MSK) provides a managed Kafka streaming service, but this adds complexity when trying to integrate and correlate logs and traces from a microservice environment. Setting up Kafka requires more development effort compared to using FluentBit and OpenTelemetry.

Option E (AWS Glue) is primarily an ETL (Extract, Transform, Load) service. While Glue is powerful for data processing, it is not a native tool for log and trace correlation, and using it would add unnecessary complexity for this use case.

Conclusion:

To meet the requirements with the least development effort:

Use FluentBit for log collection and OpenTelemetry for tracing (Option A).

Correlate logs and traces using Amazon OpenSearch (Option D).

This approach leverages AWS-native services designed for seamless integration with microservices hosted on Amazon EKS and ensures effective monitoring with minimal overhead.

The best solution for fast, event-driven processing of unpredictable file uploads is to use S3 event notifications, CloudTrail, and EventBridge to automatically trigger the AWS Glue workflow:

“You can configure S3 PutObject events to be captured by CloudTrail and forwarded through EventBridge to trigger an AWS Glue job or workflow. This allows Glue to begin processing as soon as the file arrives, with minimal latency.”

– Ace the AWS Certified Data Engineer - Associate Certification - version 2 - apple.pdf

This option provides the lowest latency and least manual overhead compared to polling or scheduling solutions.

To query the Sales table in Amazon Redshift for city names that start with " San " or " El, " the appropriate query uses a regular expression (regex) pattern to match city names that begin with those prefixes.

Option B: Select * from Sales where city_name ~ ' ^(San|El) ' ;In Amazon Redshift, the ~ operator is used to perform pattern matching using regular expressions. The ^(San|El) pattern matches city names that start with " San " or " El. " This is the correct SQL syntax for this use case.

Other options (A, C, D) contain incorrect syntax or incorrect use of special characters, making them invalid queries.

Problem Analysis:

The pipeline processes compressed files in S3 and must support incremental data processing.

AWS Glue features must facilitate tracking progress to avoid reprocessing the same data.

Key Considerations:

Incremental data processing requires tracking which files or partitions have already been processed.

The solution must be automated and efficient for large-scale ETL jobs.

Solution Analysis:

Option A: Workflows

Workflows organize and orchestrate multiple Glue jobs but do not track progress for incremental data processing.

Option B: Triggers

Triggers initiate Glue jobs based on a schedule or events but do not track which data has been processed.

Option C: Job Bookmarks

Job bookmarks track the state of the data that has been processed, enabling incremental processing.

Automatically skip files or partitions that were previously processed in Glue jobs.

Option D: Classifiers

Classifiers determine the schema of incoming data but do not handle incremental processing.

Final Recommendation:

Job bookmarks are specifically designed to enable incremental data processing in AWS Glue ETL pipelines.

AWS Glue Job Bookmarks Documentation

AWS Glue ETL Features

Option A (AWS DMS): Lets you replicate only selected tables from Aurora to Redshift. It supports ongoing replication (CDC) and reduces unnecessary data transfers.

Option C (Zero-ETL with Aurora to Redshift): This new integration allows real-time, serverless, and low-maintenance data replication. It ' s designed to reduce operational overhead drastically.

Other options introduce manual processing or inefficient query access (Option D). Glue workflows (Option E) add unnecessary complexity and lag for near-real-time needs.

“Zero-ETL integration between Amazon Aurora and Redshift enables real-time analytics without building complex pipelines.”

Problem Analysis:

The company must log all user activities and connection activities in Amazon Redshift for security compliance.

Key Considerations:

Redshift supports audit logging, which can be configured to write logs to an S3 bucket.

S3 provides durable, scalable, and cost-effective storage for logs.

Solution Analysis:

Option A: S3 for Logging

Standard approach for storing Redshift logs.

Easy to set up and manage with minimal cost.

Option B: Amazon EFS

EFS is unnecessary for this use case and less cost-efficient than S3.

Option C: Aurora MySQL

Using a database to store logs increases complexity and cost.

Option D: EBS Volume

EBS is not a scalable option for log storage compared to S3.

Final Recommendation:

Enable Redshift audit logging and specify an S3 bucket as the destination.

Amazon Redshift Audit Logging

Storing Logs in Amazon S3

Amazon RDS controls network-level access through security groups, which act as virtual firewalls. To allow an EC2 instance to connect to an RDS database using JDBC, the RDS security group must explicitly allow inbound traffic from the EC2 instance or its associated security group on the appropriate database port.

IAM roles do not control network connectivity to RDS databases. The ec2_authorized_hosts parameter does not exist for RDS engines. The Amazon RDS Data API is not supported for standard JDBC connections and is only available for Aurora Serverless v1 and v2.

Updating the RDS security group is the correct and simplest approach. By adding an inbound rule that allows traffic from the EC2 instance’s security group, the application can securely connect to the database without additional configuration or restarts.

Therefore, Option C is the correct solution.

Option A is the most operationally efficient way to meet the requirements because it minimizes the number of steps and services involved in the data export process. AWS Glue is a fully managed service that can extract, transform, and load (ETL) data from various sources to various destinations, including Amazon S3. AWS Glue can also convert data to different formats, such as Parquet, which is a columnar storage format that is optimized for analytics. By creating a view in the SQL Server databases that contains the required data elements, the AWS Glue job can select the data directly from the view without having to perform any joins or transformations on the source data. The AWS Glue job can then transfer the data in Parquet format to an S3 bucket and run on a daily schedule.

Option B is not operationally efficient because it involves multiple steps and services to export the data. SQL Server Agent is a tool that can run scheduled tasks on SQL Server databases, such as executing SQL queries. However, SQL Server Agent cannot directly export data to S3, so the query output must be saved as .csv objects on the EC2 instance. Then, an S3 event must be configured to trigger an AWS Lambda function that can transform the .csv objects to Parquet format and upload them to S3. This option adds complexity and latency to the data export process and requires additional resources and configuration.

Option C is not operationally efficient because it introduces an unnecessary step of running an AWS Glue crawler to read the view. An AWS Glue crawler is a service that can scan data sources and create metadata tables in the AWS Glue Data Catalog. The Data Catalog is a central repository that stores information about the data sources, such as schema, format, and location. However, in this scenario, the schema and format of the data elements are already known and fixed, so there is no need to run a crawler to discover them. The AWS Glue job can directly select the data from the view without using the Data Catalog. Running a crawler adds extra time and cost to the data export process.

Option D is not operationally efficient because it requires custom code and configuration to query the databases and transform the data. An AWS Lambda function is a service that can run code in response to events or triggers, such as Amazon EventBridge. Amazon EventBridge is a service that can connect applications and services with event sources, such as schedules, and route them to targets, such as Lambda functions. However, in this scenario, using a Lambda function to query the databases and transform the data is not the best option because it requires writing and maintaining code that uses JDBC to connect to the SQL Server databases, retrieve the required data, convert the data to Parquet format, and transfer the data to S3. This option also has limitations on the execution time, memory, and concurrency of the Lambda function, which may affect the performance and reliability of the data export process.

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide

AWS Glue Documentation

Working with Views in AWS Glue

Converting to Columnar Formats

 Lambda layers are a way to share code and dependencies across multiple Lambda functions. By packaging the custom Python scripts into Lambda layers, the data engineer can update the scripts in one place and have them automatically applied to all the Lambda functions that use the layer. This reduces the manual effort and ensures consistency across the Lambda functions. The other options are either not feasible or not efficient. Storing a pointer to the custom Python scripts in the execution context object or in environment variables would require the Lambda functions to download the scripts from Amazon S3 every time they are invoked, which would increase latency and cost. Assigning the same alias to each Lambda function would not help with updating the Python scripts, as the alias only points to a specific version of the Lambda function code. References:

AWS Lambda layers

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide, Chapter 3: Data Ingestion and Transformation, Section 3.4: AWS Lambda

Option C is correct because SSE-KMS encrypts Amazon S3 objects by using AWS KMS keys, and access to those keys can be controlled through IAM policies and KMS key policies so that only specific employees can use them. This directly satisfies both requirements: encryption of the S3 objects and restriction of key usage to a limited group of authorized users. AWS-managed S3 encryption with KMS also requires far less effort than building or operating a dedicated HSM-based solution. This is the standard AWS answer when the question asks for encrypted S3 data with fine-grained control over who can use the encryption keys.

Option A is far more operationally complex because CloudHSM requires managing dedicated HSM infrastructure. Option B is less desirable operationally because SSE-C requires the customer to provide and manage the keys for every request. Option D is incorrect because SSE-S3 uses Amazon S3 managed keys, and customers do not get the same granular control over key usage by specific employees. Therefore, the least-effort and best-governed solution is SSE-KMS with restricted KMS key access. This aligns with the study guide’s security-and-governance focus on using AWS KMS when controlled access to encryption keys is required.

Amazon RDS is a fully managed service that provides relational databases in the cloud. Amazon RDS for MySQL is one of the supported database engines that you can use to run your applications. Amazon RDS provides various features and tools to monitor and optimize the performance of your DB instances, such as Performance Insights, Enhanced Monitoring, CloudWatch metrics and alarms, etc.

Using the Performance Insights feature of Amazon RDS to identify queries that have high CPU utilization and optimizing the problematic queries will help reduce the CPU utilization of the DB instance. Performance Insights is a feature that allows you to analyze the load on your DB instance and determine what is causing performance issues. Performance Insights collects, analyzes, and displays database performance data using an interactive dashboard. You can use Performance Insights to identify the top SQL statements, hosts, users, or processes that are consuming the most CPU resources. You can also drill down into the details of each query and see the execution plan, wait events, locks, etc. By using Performance Insights, you can pinpoint the root cause of the high CPU utilization and optimize the queries accordingly. For example, you can rewrite the queries to make them more efficient, add or remove indexes, use prepared statements, etc.

Implementing caching to reduce the database query load will also help reduce the CPU utilization of the DB instance. Caching is a technique that allows you to store frequently accessed data in a fast and scalable storage layer, such as Amazon ElastiCache. By using caching, you can reduce the number of requests that hit your database, which in turn reduces the CPU load on your DB instance. Caching also improves the performance and availability of your application, as it reduces the latency and increases the throughput of your data access. You can use caching for various scenarios, such as storing session data, user preferences, application configuration, etc. You can also use caching for read-heavy workloads, such as displaying product details, recommendations, reviews, etc.

The other options are not as effective as using Performance Insights and caching. Modifying the database schema to include additional tables and indexes may or may not improve the CPU utilization, depending on the nature of the workload and the queries. Adding more tables and indexes may increase the complexity and overhead of the database, which may negatively affect the performance. Rebooting the RDS DB instance once each week will not reduce the CPU utilization, as it will not address the underlying cause of the high CPU load. Rebooting may also cause downtime and disruption to your application. Upgrading to a larger instance size may reduce the CPU utilization, but it will also increase the cost and complexity of your solution. Upgrading may also not be necessary if you can optimize the queries and reduce the database load by using caching. References:

Amazon RDS

Performance Insights

Amazon ElastiCache

[AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide], Chapter 3: Data Storage and Management, Section 3.1: Amazon RDS

To achieve the most cost-effective storage solution, the data engineer needs to use an S3 Lifecycle policy that transitions objects to lower-cost storage classes based on their access patterns, and deletes them when they are no longer needed. The storage classes should also provide high availability, which means they should be resilient to the loss of data in a single Availability Zone1. Therefore, the solution must include the following steps:

Transition objects to S3 Standard-Infrequent Access (S3 Standard-IA) after 6 months. S3 Standard-IA is designed for data that is accessed less frequently, but requires rapid access when needed. It offers the same high durability, throughput, and low latency as S3 Standard, but with a lower storage cost and a retrieval fee2. Therefore, it is suitable for data files that are accessed once or twice each month. S3 Standard-IA also provides high availability, as it stores data redundantly across multiple Availability Zones1.

Transfer objects to S3 Glacier Deep Archive after 2 years. S3 Glacier Deep Archive is the lowest-cost storage class that offers secure and durable storage for data that is rarely accessed and can tolerate a 12-hour retrieval time. It is ideal for long-term archiving and digital preservation3. Therefore, it is suitable for data files that are accessed only once or twice each year. S3 Glacier Deep Archive also provides high availability, as it stores data across at least three geographically dispersed Availability Zones1.

Delete objects when they are no longer needed. The data engineer can specify an expiration action in the S3 Lifecycle policy to delete objects after a certain period of time. This will reduce the storage cost and comply with any data retention policies.

Option C is the only solution that includes all these steps. Therefore, option C is the correct answer.

Option A is incorrect because it transitions objects to S3 One Zone-Infrequent Access (S3 One Zone-IA) after 6 months. S3 One Zone-IA is similar to S3 Standard-IA, but it stores data in a single Availability Zone. This means it has a lower availability and durability than S3 Standard-IA, and it is not resilient to the loss of data in a single Availability Zone1. Therefore, it does not provide high availability as required.

Option B is incorrect because it transfers objects to S3 Glacier Flexible Retrieval after 2 years. S3 Glacier Flexible Retrieval is a storage class that offers secure and durable storage for data that is accessed infrequently and can tolerate a retrieval time of minutes to hours. It is more expensive than S3 Glacier Deep Archive, and it is not suitable for data that is accessed only once or twice each year3. Therefore, it is not the most cost-effective option.

Option D is incorrect because it combines the errors of option A and B. It transitions objects to S3 One Zone-IA after 6 months, which does not provide high availability, and it transfers objects to S3 Glacier Flexible Retrieval after 2 years, which is not the most cost-effective option.

1: Amazon S3 storage classes - Amazon Simple Storage Service

2: Amazon S3 Standard-Infrequent Access (S3 Standard-IA) - Amazon Simple Storage Service

3: Amazon S3 Glacier and S3 Glacier Deep Archive - Amazon Simple Storage Service

[4]: Expiring objects - Amazon Simple Storage Service

[5]: Managing your storage lifecycle - Amazon Simple Storage Service

[6]: Examples of S3 Lifecycle configuration - Amazon Simple Storage Service

[7]: Amazon S3 Lifecycle further optimizes storage cost savings with new features - What’s New with AWS

Option B is correct because the requirement is to retain a very large dataset indefinitely and analyze it from Amazon Redshift in the most cost-effective way. Amazon Redshift Spectrum allows Redshift to query data directly in Amazon S3 without loading all of the data into Redshift-managed storage. That reduces warehouse storage cost for long-term retained data. AWS documentation also recommends using Apache Parquet for Spectrum because Parquet is a columnar format, which allows Redshift Spectrum to read only the columns needed instead of scanning entire text files. This improves performance and lowers query cost.

Option A and D are less cost-effective because auto-copy loads data into the Redshift cluster, which means the company pays to store all historical data in Redshift even though the requirement is indefinite retention of a large volume of IoT data. Option C is worse than B because JSON is a row-oriented text format, and AWS guidance says columnar formats such as Parquet or ORC are preferred for Redshift Spectrum for better scan efficiency and lower cost.

 AWS Glue workflows are a cost-effective way to orchestrate complex ETL jobs that involve multiple crawlers, jobs, and triggers. AWS Glue workflows allow you to visually monitor the progress and dependencies of your ETL tasks, and automatically handle errors and retries. AWS Glue workflows also integrate with other AWS services, such as Amazon S3, Amazon Redshift, and AWS Lambda, among others, enabling you to leverage these services for your data processing workflows. AWS Glue workflows are serverless, meaning you only pay for the resources you use, and you don’t have to manage any infrastructure.

AWS Step Functions, AWS Glue Studio, and Amazon MWAA are also possible options for orchestrating ETL pipelines, but they have some drawbacks compared to AWS Glue workflows. AWS Step Functions is a serverless function orchestrator that can handle different types of data processing, such as real-time, batch, and stream processing. However, AWS Step Functions requires you to write code to define your state machines, which can be complex and error-prone. AWS Step Functions also charges you for every state transition, which can add up quickly for large-scale ETL pipelines.

AWS Glue Studio is a graphical interface that allows you to create and run AWS Glue ETL jobs without writing code. AWS Glue Studio simplifies the process of building, debugging, and monitoring your ETL jobs, and provides a range of pre-built transformations and connectors. However, AWS Glue Studio does not support workflows, meaning you cannot orchestrate multiple ETL jobs or crawlers with dependencies and triggers. AWS Glue Studio also does not support streaming data sources or targets, which limits its use cases for real-time data processing.

Amazon MWAA is a fully managed service that makes it easy to run open-source versions of Apache Airflow on AWS and build workflows to run your ETL jobs and data pipelines. Amazon MWAA provides a familiar and flexible environment for data engineers who are familiar with Apache Airflow, and integrates with a range of AWS services such as Amazon EMR, AWS Glue, and AWS Step Functions. However, Amazon MWAA is not serverless, meaning you have to provision and pay for the resources you need, regardless of your usage. Amazon MWAA also requires you to write code to define your DAGs, which can be challenging and time-consuming for complex ETL pipelines. References:

AWS Glue Workflows

AWS Step Functions

AWS Glue Studio

Amazon MWAA

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide

Lake Formation requires that S3 locations be registered before they can be accessed by AWS Glue crawlers, jobs, or queries. Registering the location associates it with the Lake Formation permissions framework.

“Before AWS Glue crawlers can access an S3 path managed by Lake Formation, the S3 location must be registered in Lake Formation.”

– Ace the AWS Certified Data Engineer - Associate Certification - version 2 - apple.pdf

The STL ALERT EVENT LOG table view records anomalies when the query optimizer identifies conditions that might indicate performance issues. These conditions include skewed data distribution, missing statistics, nested loop joins, and broadcasted data. The STL ALERT EVENT LOG table view can help the data engineer to identify and troubleshoot the root causes of performance issues and optimize the query execution plan. The other table views are not relevant for this requirement. STL USAGE CONTROL records the usage limits and quotas for Amazon Redshift resources. STL QUERY METRICS records the execution time and resource consumption of queries. STL PLAN INFO records the query execution plan and the steps involved in each query. References:

STL ALERT EVENT LOG

System Tables and Views

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide

Amazon Kinesis Data Streams is a service that enables you to collect, process, and analyze streaming data in real time. You can use Kinesis Data Streams to capture sensor data from various sources, such as IoT devices, web applications, or mobile apps. You can create data streams that can scale up to handle any amount of data from thousands of producers. You can also use the Kinesis Client Library (KCL) or the Kinesis Data Streams API to write applications that process and analyze the data in the streams1.

Amazon DynamoDB is a fully managed NoSQL database service that provides fast and predictable performance with seamless scalability. You can use DynamoDB to store the sensor data in nested JSON format, as DynamoDB supports document data types, such as lists and maps. You can also use DynamoDB to query the data with a latency of less than 10 milliseconds, as DynamoDB offers single-digit millisecond performance for any scale of data. You can use the DynamoDB API or the AWS SDKs to perform queries on the data, such as using key-value lookups, scans, or queries2.

The solution that meets the requirements with the least operational overhead is to use Amazon Kinesis Data Streams to capture the sensor data and store the data in Amazon DynamoDB for querying. This solution has the following advantages:

It does not require you to provision, manage, or scale any servers, clusters, or queues, as Kinesis Data Streams and DynamoDB are fully managed services that handle all the infrastructure for you. This reduces the operational complexity and cost of running your solution.

It allows you to ingest sensor data in near real time, as Kinesis Data Streams can capture data records as they are produced and deliver them to your applications within seconds. You can also use Kinesis Data Firehose to load the data from the streams to DynamoDB automatically and continuously3.

It allows you to store the data in nested JSON format, as DynamoDB supports document data types, such as lists and maps. You can also use DynamoDB Streams to capture changes in the data and trigger actions, such as sending notifications or updating other databases.

It allows you to query the data with a latency of less than 10 milliseconds, as DynamoDB offers single-digit millisecond performance for any scale of data. You can also use DynamoDB Accelerator (DAX) to improve the read performance by caching frequently accessed data.

Option A is incorrect because it suggests using a self-hosted Apache Kafka cluster to capture the sensor data and store the data in Amazon S3 for querying. This solution has the following disadvantages:

It requires you to provision, manage, and scale your own Kafka cluster, either on EC2 instances or on-premises servers. This increases the operational complexity and cost of running your solution.

It does not allow you to query the data with a latency of less than 10 milliseconds, as Amazon S3 is an object storage service that is not optimized for low-latency queries. You need to use another service, such as Amazon Athena or Amazon Redshift Spectrum, to query the data in S3, which may incur additional costs and latency.

Option B is incorrect because it suggests using AWS Lambda to process the sensor data and store the data in Amazon S3 for querying. This solution has the following disadvantages:

It does not allow you to ingest sensor data in near real time, as Lambda is a serverless compute service that runs code in response to events. You need to use another service, such as API Gateway or Kinesis Data Streams, to trigger Lambda functions with sensor data, which may add extra latency and complexity to your solution.

It does not allow you to query the data with a latency of less than 10 milliseconds, as Amazon S3 is an object storage service that is not optimized for low-latency queries. You need to use another service, such as Amazon Athena or Amazon Redshift Spectrum, to query the data in S3, which may incur additional costs and latency.

Option D is incorrect because it suggests using Amazon Simple Queue Service (Amazon SQS) to buffer incoming sensor data and use AWS Glue to store the data in Amazon RDS for querying. This solution has the following disadvantages:

It does not allow you to ingest sensor data in near real time, as Amazon SQS is a message queue service that delivers messages in a best-effort manner. You need to use another service, such as Lambda or EC2, to poll the messages from the queue and process them, which may add extra latency and complexity to your solution.

It does not allow you to store the data in nested JSON format, as Amazon RDS is a relational database service that supports structured data types, such as tables and columns. You need to use another service, such as AWS Glue, to transform the data from JSON to relational format, which may add extra cost and overhead to your solution.

1: Amazon Kinesis Data Streams - Features

2: Amazon DynamoDB - Features

3: Loading Streaming Data into Amazon DynamoDB - Amazon Kinesis Data Firehose

[4]: Capturing Table Activity with DynamoDB Streams - Amazon DynamoDB

[5]: Amazon DynamoDB Accelerator (DAX) - Features

[6]: Amazon S3 - Features

[7]: AWS Lambda - Features

[8]: Amazon Simple Queue Service - Features

[9]: Amazon Relational Database Service - Features

[10]: Working with JSON in Amazon RDS - Amazon Relational Database Service

[11]: AWS Glue - Features

The correct answer is B because Apache Iceberg is designed for analytical data lake tables that need schema evolution, transactional consistency, and historical snapshot access. AWS Glue supports Apache Iceberg for data lakes and can read and write Iceberg tables in Amazon S3 through the AWS Glue Data Catalog. The AWS Glue documentation also shows Iceberg tables being created with TBLPROPERTIES ( " format-version " = " 2 " ), which is the correct table format direction for modern Iceberg capabilities. CSV with scheduled conversion does not provide ACID transactions or point-in-time table snapshots. S3 Versioning tracks object versions but does not provide table-level ACID semantics. DynamoDB is not the best fit for an S3-based analytical data lake requirement.

Option A is correct because Amazon Kinesis Data Streams is designed to collect and process large streams of data records in real time and is optimized for rapid, continuous data intake and aggregation. AWS states that Kinesis Data Streams is a fully managed, serverless data streaming service for ingesting streaming data at any scale and highlights event sources such as clickstream and application activity data. AWS also notes that Kinesis can pair with AWS Lambda for low-latency event processing. This directly matches the requirement to ingest millions of customer activity events with minimum latency in a scalable and durable way.

Option B is batch processing every 10 minutes, so it does not satisfy the real-time requirement. Option C depends on files landing in S3 first, which is also not the lowest-latency ingest path. Option D can support streaming, but a managed Kafka cluster with self-managed consumers introduces more operational overhead and is generally less cost-effective here than a native Kinesis + Lambda pipeline for an AWS-centric event-ingestion workload. Therefore, Kinesis Data Streams + Lambda + Redshift is the best answer.

 Amazon DynamoDB is a fully managed NoSQL database service that provides fast and predictable performance with seamless scalability. DynamoDB offers two capacity modes for throughput capacity: provisioned and on-demand. In provisioned capacity mode, you specify the number of read and write capacity units per second that you expect your application to require. DynamoDB reserves the resources to meet your throughput needs with consistent performance. In on-demand capacity mode, you pay per request and DynamoDB scales the resources up and down automatically based on the actual workload. On-demand capacity mode is suitable for unpredictable workloads that can vary significantly over time1.

The solution that meets the requirements in the most cost-effective way is to use AWS Application Auto Scaling to schedule higher provisioned capacity for peak usage times and lower capacity during off-peak times. This solution has the following advantages:

It allows you to optimize the cost and performance of your DynamoDB table by adjusting the provisioned capacity according to your predictable workload patterns. You can use scheduled scaling to specify the date and time for the scaling actions, and the new minimum and maximum capacity limits. For example, you can schedule higher capacity for every Monday morning and lower capacity for weekends2.

It enables you to take advantage of the lower cost per unit of provisioned capacity mode compared to on-demand capacity mode. Provisioned capacity mode charges a flat hourly rate for the capacity you reserve, regardless of how much you use. On-demand capacity mode charges for each read and write request you consume, with no minimum capacity required. For predictable workloads, provisioned capacity mode can be more cost-effective than on-demand capacity mode1.

It ensures that your application performs consistently during peak usage times by having enough capacity to handle the increased load. You can also use auto scaling to automatically adjust the provisioned capacity based on the actual utilization of your table, and set a target utilization percentage for your table or global secondary index. This way, you can avoid under-provisioning or over-provisioning your table2.

Option A is incorrect because it suggests increasing the provisioned capacity to the maximum capacity that is currently present during peak load times. This solution has the following disadvantages:

It wastes money by paying for unused capacity during off-peak times. If you provision the same high capacity for all times, regardless of the actual workload, you are over-provisioning your table and paying for resources that you don’t need1.

It does not account for possible changes in the workload patterns over time. If your peak load times increase or decrease in the future, you may need to manually adjust the provisioned capacity to match the new demand. This adds operational overhead and complexity to your application2.

Option B is incorrect because it suggests dividing the table into two tables and provisioning each table with half of the provisioned capacity of the original table. This solution has the following disadvantages:

It complicates the data model and the application logic by splitting the data into two separate tables. You need to ensure that the queries are evenly distributed across both tables, and that the data is consistent and synchronized between them. This adds extra development and maintenance effort to your application3.

It does not solve the problem of adjusting the provisioned capacity according to the workload patterns. You still need to manually or automatically scale the capacity of each table based on the actual utilization and demand. This may result in under-provisioning or over-provisioning your tables2.

Option D is incorrect because it suggests changing the capacity mode from provisioned to on-demand. This solution has the following disadvantages:

It may incur higher costs than provisioned capacity mode for predictable workloads. On-demand capacity mode charges for each read and write request you consume, with no minimum capacity required. For predictable workloads, provisioned capacity mode can be more cost-effective than on-demand capacity mode, as you can reserve the capacity you need at a lower rate1.

It may not provide consistent performance during peak usage times, as on-demand capacity mode may take some time to scale up the resources to meet the sudden increase in demand. On-demand capacity mode uses adaptive capacity to handle bursts of traffic, but it may not be able to handle very large spikes or sustained high throughput. In such cases, you may experience throttling or increased latency.

1: Choosing the right DynamoDB capacity mode - Amazon DynamoDB

2: Managing throughput capacity automatically with DynamoDB auto scaling - Amazon DynamoDB

3: Best practices for designing and using partition keys effectively - Amazon DynamoDB

[4]: On-demand mode guidelines - Amazon DynamoDB

[5]: How to optimize Amazon DynamoDB costs - AWS Database Blog

[6]: DynamoDB adaptive capacity: How it works and how it helps - AWS Database Blog

[7]: Amazon DynamoDB pricing - Amazon Web Services (AWS)

The company wants to ensure that no customer records are deleted or modified for 7 years, and even the root user should not have the ability to change the data. S3 Object Lock in Compliance Mode is the correct solution for this scenario.

Option B: Enable compliance mode on the S3 bucket. Use a default retention period of 7 years.In Compliance Mode, even the root user cannot delete or modify locked objects during the retention period. This ensures that the data is protected for the entire 7-year duration as required. Compliance mode is stricter than governance mode and prevents all forms of alteration, even by privileged users.

Option A (Governance Mode) still allows certain privileged users (like the root user) to bypass the lock, which does not meet the company ' s requirement. Option C (legal hold) and Option D (setting retention per object) do not fully address the requirement to block root user modifications.

Option D is the best solution to meet the requirements with the least operational overhead because AWS Lake Formation is a fully managed service that simplifies the process of building, securing, and managing data lakes. AWS Lake Formation allows you to define granular data access policies at the row and column level for different users and groups. AWS Lake Formation also integrates with Amazon Athena, Amazon Redshift Spectrum, and Apache Hive on Amazon EMR, enabling these services to access the data in the data lake through AWS Lake Formation.

Option A is not a good solution because S3 access policies cannot restrict data access by rows and columns. S3 access policies are based on the identity and permissions of the requester, the bucket and object ownership, and the object prefix and tags. S3 access policies cannot enforce fine-grained data access control at the row and column level.

Option B is not a good solution because it involves using Apache Ranger and Apache Pig, which are not fully managed services and require additional configuration and maintenance. Apache Ranger is a framework that provides centralized security administration for data stored in Hadoop clusters, such as Amazon EMR. Apache Ranger can enforce row-level and column-level access policies for Apache Hive tables. However, Apache Ranger is not a native AWS service and requires manual installation and configuration on Amazon EMR clusters. Apache Pig is a platform that allows you to analyze large data sets using a high-level scripting language called Pig Latin. Apache Pig can access data stored in Amazon S3 and process it using Apache Hive. However, Apache Pig is not a native AWS service and requires manual installation and configuration on Amazon EMR clusters.

Option C is not a good solution because Amazon Redshift is not a suitable service for data lake storage. Amazon Redshift is a fully managed data warehouse service that allows you to run complex analytical queries using standard SQL. Amazon Redshift can enforce row-level and column-level access policies for different users and groups. However, Amazon Redshift is not designed to store and process large volumes of unstructured or semi-structured data, which are typical characteristics of data lakes. Amazon Redshift is also more expensive and less scalable than Amazon S3 for data lake storage.

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide

What Is AWS Lake Formation? - AWS Lake Formation

Using AWS Lake Formation with Amazon Athena - AWS Lake Formation

Using AWS Lake Formation with Amazon Redshift Spectrum - AWS Lake Formation

Using AWS Lake Formation with Apache Hive on Amazon EMR - AWS Lake Formation

Using Bucket Policies and User Policies - Amazon Simple Storage Service

Apache Ranger

Apache Pig

What Is Amazon Redshift? - Amazon Redshift

This solution meets the requirements of managing the ingestion of real-time streaming data into AWS and performing real-time analytics on the incoming streaming data with the least operational overhead. Amazon Managed Service for Apache Flink is a fully managed service that allows you to run Apache Flink applications without having to manage any infrastructure or clusters. Apache Flink is a framework for stateful stream processing that supports various types of aggregations, such as tumbling, sliding, and session windows, over streaming data. By using Amazon Managed Service for Apache Flink, you can easily connect to Amazon Kinesis Data Streams as the source and sink of your streaming data, and perform time-based analytics over a window of up to 30 minutes. This solution is also highly fault tolerant, as Amazon Managed Service for Apache Flink automatically scales, monitors, and restarts your Flink applications in case of failures. References:

Amazon Managed Service for Apache Flink

Apache Flink

Window Aggregations in Flink

Amazon Redshift is a fully managed, petabyte-scale data warehouse service that enables fast and cost-effective analysis of large volumes of data. Amazon Redshift uses columnar storage, compression, and zone maps to optimize the storage and performance of data. However, over time, as data is inserted, updated, or deleted, the physical storage of data can become fragmented, resulting in wasted disk space and degraded query performance. To address this issue, Amazon Redshift provides the VACUUM command, which reclaims disk space and resorts rows in either a specified table or all tables in the current schema1.

The VACUUM command has four options: FULL, DELETE ONLY, SORT ONLY, and REINDEX. The option that best meets the requirements of the question is VACUUM REINDEX, which re-sorts the rows in a table that has an interleaved sort key and rewrites the table to a new location on disk. An interleaved sort key is a type of sort key that gives equal weight to each column in the sort key, and stores the rows in a way that optimizes the performance of queries that filter by multiple columns in the sort key. However, as data is added or changed, the interleaved sort order can become skewed, resulting in suboptimal query performance. The VACUUM REINDEX option restores the optimal interleaved sort order and reclaims disk space by removing deleted rows. This option also analyzes the sort key column and updates the table statistics, which are used by the query optimizer to generate the most efficient query execution plan23.

The other options are not optimal for the following reasons:

A. VACUUM FULL Orders. This option reclaims disk space by removing deleted rows and resorts the entire table. However, this option is not suitable for tables that have an interleaved sort key, as it does not restore the optimal interleaved sort order. Moreover, this option is the most resource-intensive and time-consuming, as it rewrites the entire table to a new location on disk.

B. VACUUM DELETE ONLY Orders. This option reclaims disk space by removing deleted rows, but does not resort the table. This option is not suitable for tables that have any sort key, as it does not improve the query performance by restoring the sort order. Moreover, this option does not analyze the sort key column and update the table statistics.

D. VACUUM SORT ONLY Orders. This option resorts the entire table, but does not reclaim disk space by removing deleted rows. This option is not suitable for tables that have an interleaved sort key, as it does not restore the optimal interleaved sort order. Moreover, this option does not analyze the sort key column and update the table statistics.

1: Amazon Redshift VACUUM

2: Amazon Redshift Interleaved Sorting

3: Amazon Redshift ANALYZE

Option C is correct because Amazon OpenSearch Service is designed for search and analytics in near real time with low-latency queries. AWS documentation states that Amazon OpenSearch Service makes it easy to deploy, secure, operate, and scale OpenSearch to search, analyze, and visualize data in real time, and that it provides real-time analytics for use cases such as log analytics, full-text search, application monitoring, and clickstream analytics. That aligns directly with the requirement for handling many requests per second with low query latency.

This option also best satisfies the requirement that individual data teams can own and configure their own environments for cost and performance optimization. By assigning each team a separate OpenSearch cluster, each team can independently tune indexing, storage, shard layout, retention, and scaling policies. Option A is less suitable because Amazon Athena is serverless SQL over data in S3 and is not the best fit for low-latency, high-request-rate interactive search workloads. Option B focuses on streaming data processing, not primary low-latency indexed search. Option D uses a relational database, which is not the native AWS choice for large-scale search and analytics. Therefore, OpenSearch Service is the most appropriate and scalable solution.

 Amazon Athena is a serverless interactive query service that allows you to analyze data in Amazon S3 using standard SQL. Athena supports various data formats, such as CSV, JSON, ORC, Avro, and Parquet. However, not all data formats are equally efficient for querying. Some data formats, such as CSV and JSON, are row-oriented, meaning that they store data as a sequence of records, each with the same fields. Row-oriented formats are suitable for loading and exporting data, but they are not optimal for analytical queries that often access only a subset of columns. Row-oriented formats also do not support compression or encoding techniques that can reduce the data size and improve the query performance.

On the other hand, some data formats, such as ORC and Parquet, are column-oriented, meaning that they store data as a collection of columns, each with a specific data type. Column-oriented formats are ideal for analytical queries that often filter, aggregate, or join data by columns. Column-oriented formats also support compression and encoding techniques that can reduce the data size and improve the query performance. For example, Parquet supports dictionary encoding, which replaces repeated values with numeric codes, and run-length encoding, which replaces consecutive identical values with a single value and a count. Parquet also supports various compression algorithms, such as Snappy, GZIP, and ZSTD, that can further reduce the data size and improve the query performance.

Therefore, changing the data format from CSV to Parquet and applying Snappy compression will most speed up the Athena query performance. Parquet is a column-oriented format that allows Athena to scan only the relevant columns and skip the rest, reducing the amount of data read from S3. Snappy is a compression algorithm that reduces the data size without compromising the query speed, as it is splittable and does not require decompression before reading. This solution will also reduce the cost of Athena queries, as Athena charges based on the amount of data scanned from S3.

The other options are not as effective as changing the data format to Parquet and applying Snappy compression. Changing the data format from CSV to JSON and applying Snappy compression will not improve the query performance significantly, as JSON is also a row-oriented format that does not support columnar access or encoding techniques. Compressing the CSV files by using Snappy compression will reduce the data size, but it will not improve the query performance significantly, as CSV is still a row-oriented format that does not support columnar access or encoding techniques. Compressing the CSV files by using gzjg compression will reduce the data size, but it will degrade the query performance, as gzjg is not a splittable compression algorithm and requires decompression before reading. References:

Amazon Athena

Choosing the Right Data Format

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide, Chapter 5: Data Analysis and Visualization, Section 5.1: Amazon Athena

AWS Database Migration Service (AWS DMS) is purpose-built to migrate and continuously replicate data from both AWS-hosted and on-premises databases. It supports full-load (existing data) and change data capture (CDC) for ongoing changes, making it the most cost-effective and operationally simple solution in this scenario.

“DMS supports both full-load and continuous replication via CDC. This enables replicating existing and future data from various sources to a data lake in Amazon S3.”

– Ace the AWS Certified Data Engineer - Associate Certification - version 2 - apple.pdf

AWS Glue is not suitable for real-time CDC replication across hybrid environments.

The issue described is that the AWS Glue job is reprocessing files from previous runs despite the bookmark feature being enabled. Bookmarks in AWS Glue allow jobs to keep track of which files or data have already been processed to avoid reprocessing. The most likely reason for reprocessing the files is missing S3 permissions, specifically s3

s3

is a permission required by AWS Glue when bookmarks are enabled to ensure Glue can retrieve metadata from the files in S3, which is necessary for the bookmark mechanism to function correctly. Without this permission, Glue cannot track which files have been processed, resulting in reprocessing during subsequent runs.

Concurrency settings (Option B) and the version of AWS Glue (Option C) do not affect the bookmark behavior. Similarly, the lack of a commit statement (Option D) is not applicable in this context, as Glue handles commits internally when interacting with Redshift and S3.

Thus, the root cause is likely related to insufficient permissions on the S3 bucket, specifically s3

, which is required for bookmarks to work as expected.

The FLEX execution class allows you to run AWS Glue jobs on spare compute capacity instead of dedicated hardware. This can reduce the cost of running non-urgent or non-time sensitive data integration workloads, such as testing and one-time data loads. The FLEX execution class is available for AWS Glue 3.0 Spark jobs. The other options are not as cost-effective as FLEX, because they either use dedicated resources (STANDARD) or do not affect the cost at all (Spot Instance type and GlueVersion). References:

Introducing AWS Glue Flex jobs: Cost savings on ETL workloads

Serverless Data Integration – AWS Glue Pricing

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide (Chapter 5, page 125)

Option B is the right approach because the requirement is to mask PII in the log destination (the CloudWatch Logs log group). The exam guide explicitly calls out security responsibilities that include “data encryption and masking” and also emphasizes enabling and preparing logs for audit and governance needs. A log-group–level masking mechanism is therefore the most direct control point to prevent sensitive values from being exposed to anyone who can view logs.

Option A (Glue security configuration) is primarily used to apply protections such as encryption settings for Glue jobs and related outputs; it does not inherently solve the problem of masking PII that has already been emitted into application logs. Option C (Macie) is intended to discover and classify sensitive data—most commonly in Amazon S3—and to produce findings; it is not a log-masking control for CloudWatch Logs. The material reinforces Macie as a discovery/classification service for PII rather than a masking mechanism. Option D adds custom code paths in the ETL job and still may miss PII that appears in framework/system logs; it also violates the “least operational effort” spirit compared to applying a centralized log policy.

Therefore, attaching a data protection policy directly to the CloudWatch Logs log group best meets the masking requirement.

The data engineer needs to orchestrate ETL jobs that include Spark jobs on Amazon EMR, API calls to Salesforce, and loading data into Redshift. They also need automatic failure handling and retries. Amazon Managed Workflows for Apache Airflow (Amazon MWAA) is the best solution for this requirement.

Option A: Amazon Managed Workflows for Apache Airflow (Amazon MWAA)Apache Airflow is designed for complex job orchestration, allowing users to define workflows (DAGs) in Python. MWAA manages Airflow and its integrations with other AWS services, including Amazon EMR, Redshift, and external APIs like Salesforce. It provides automatic retry handling, failure detection, and detailed monitoring, which fits the use case perfectly.

Option B (AWS Step Functions) can orchestrate tasks but doesn ' t natively support complex workflow definitions with Python like Airflow does.

Option C (AWS Glue) is more focused on ETL and doesn ' t handle the orchestration of external systems like Salesforce as well as Airflow.

Option D (Amazon EventBridge) is more suited for event-driven architectures rather than complex workflow orchestration.

For high-performance analytics in Athena and EMR Spark, the columnar storage formats Parquet and ORC deliver the shortest query times. These formats minimize I/O, enable predicate pushdown, and support efficient compression and parallel reads.

When integrated with the AWS Glue Data Catalog, schema evolution is automatically tracked and managed across queries.

This combination (Parquet/ORC + Glue Catalog) is explicitly recommended by AWS for analytical workloads on S3 data lakes.

“For analytical workloads using Amazon EMR and Athena, store data in columnar formats such as Parquet or ORC. These formats provide optimized compression and query performance.”

– Ace the AWS Certified Data Engineer – Associate Certification – version 2 – apple.pdf

“Use the AWS Glue Data Catalog to manage table schemas and schema evolution for columnar data formats such as Parquet and ORC.”

– Ace the AWS Certified Data Engineer – Associate Certification – version 2 – apple.pdf

Therefore, using ORC or Parquet along with the Glue Data Catalog achieves the shortest Athena query times and seamless schema management.

 Amazon Redshift Serverless is a new feature of Amazon Redshift that enables you to run SQL queries on data in Amazon S3 without provisioning or managing any clusters. You can use Amazon Redshift Serverless to automatically process the analytics workload, as it scales up and down the compute resources based on the query demand, and charges you only for the resources consumed. This solution will meet the requirements with the least operational overhead, as it does not require the data engineer to create, delete, pause, or resume any Redshift clusters, or to manage any infrastructure manually. You can use the Amazon Redshift Data API to run queries from the AWS CLI, AWS SDK, or AWS Lambda functions12.

The other options are not optimal for the following reasons:

A. Use Amazon Step Functions to pause the Redshift cluster when the analytics processes are complete and to resume the cluster to run new processes every month. This option is not recommended, as it would still require the data engineer to create and delete a new Redshift provisioned cluster every month, which can incur additional costs and time. Moreover, this option would require the data engineer to use Amazon Step Functions to orchestrate the workflow of pausing and resuming the cluster, which can add complexity and overhead.

C. Use the AWS CLI to automatically process the analytics workload. This option is vague and does not specify how the AWS CLI is used to process the analytics workload. The AWS CLI can be used to run queries on data in Amazon S3 using Amazon Redshift Serverless, Amazon Athena, or Amazon EMR, but each of these services has different features and benefits. Moreover, this option does not address the requirement of not managing the infrastructure manually, as the data engineer may still need to provision and configure some resources, such as Amazon EMR clusters or Amazon Athena workgroups.

D. Use AWS CloudFormation templates to automatically process the analytics workload. This option is also vague and does not specify how AWS CloudFormation templates are used to process the analytics workload. AWS CloudFormation is a service that lets you model and provision AWS resources using templates. You can use AWS CloudFormation templates to create and delete a Redshift provisioned cluster every month, or to create and configure other AWS resources, such as Amazon EMR, Amazon Athena, or Amazon Redshift Serverless. However, this option does not address the requirement of not managing the infrastructure manually, as the data engineer may still need to write and maintain the AWS CloudFormation templates, and to monitor the status and performance of the resources.

1: Amazon Redshift Serverless

2: Amazon Redshift Data API

Amazon Step Functions

AWS CLI

AWS CloudFormation

The performance issue in Amazon Redshift Spectrum queries arises due to the nature of CSV files, which are row-based storage formats. Spectrum is more optimized for columnar formats, which significantly improve performance by reducing the amount of data scanned. Also, partitioning data based on relevant columns like order date can further reduce the amount of data scanned, as queries can focus only on the necessary partitions.

A. Configure the third-party application to create the files in a columnar format:

Columnar formats (like Parquet or ORC) store data in a way that is optimized for analytical queries because they allow queries to scan only the columns required, rather than scanning all columns in a row-based format like CSV.

Amazon Redshift Spectrum works much more efficiently with columnar formats, reducing the amount of data that needs to be scanned, which improves query performance.

Option A is correct because Amazon Redshift supports the EXCLUDE clause in the SELECT list for exactly this use case: returning all columns from a wide table except a small set of unwanted columns. AWS documentation states that “EXCLUDE column_list names the columns that are excluded from the query results” and notes that the option is helpful when only a subset of columns must be excluded from a wide table. That matches this question precisely because the table contains 100 columns and only company_id and unique_system_id need to be omitted.

Option B is invalid SQL syntax in Amazon Redshift because NOT IN is used in predicates, not in the select list to remove columns. Option C is also incorrect because EXCEPT in Redshift is a set operator used between query result sets, not a select-list column exclusion feature. AWS documents EXCEPT alongside UNION and INTERSECT as result-set comparison operators, not as a way to omit columns from SELECT *. Option D is not valid Redshift SQL syntax for column selection. Therefore, the correct and documented Redshift statement is SELECT * EXCLUDE company_id, unique_system_id FROM orders.complete_orders_history;

For Amazon QuickSight to access data in an Amazon S3 bucket that is encrypted with an AWS KMS key in a different AWS account, two distinct permissions are required: access to the S3 bucket and access to the KMS key.

First, the QuickSight service role in the BI-Account must be granted permission to access the S3 bucket. This is accomplished by adding the S3 bucket as a resource that the QuickSight service role can access and configuring the bucket policy in the Hub-Account to trust that role. Without explicit S3 permissions, QuickSight cannot read the objects.

Second, because the S3 bucket uses SSE-KMS encryption, QuickSight must also be authorized to use the KMS key. This requires adding an IAM policy to the QuickSight service role that allows kms:Decrypt and related permissions on the KMS key. Without KMS permissions, S3 access will fail even if the bucket policy allows access.

AWS RAM cannot be used to share S3 buckets, encryption of network connections is irrelevant to KMS permissions, and adding the KMS key as a “resource” alone is not sufficient without IAM permissions.

Therefore, Options B and D are correct.

The correct answer is C because SNS Publish and PublishBatch operations are data-plane operations, not ordinary management events. AWS CloudTrail documentation specifically lists Amazon SNS Publish and PublishBatch API operations on topics as supported data events. Because the company already has a CloudTrail Lake event data store, the lowest-operational-overhead approach is to enable the relevant SNS data events and query them directly in CloudTrail Lake. Creating Glue Data Catalog tables and querying with Athena adds unnecessary cataloging and query infrastructure. Management events would not satisfy the requirement to audit the requested SNS publishing actions. Therefore, the precise configuration is SNS data events plus CloudTrail Lake querying.

Option A is the correct answer because it meets the requirements with the least operational overhead. Creating an S3 event notification that has an event type of s3:ObjectCreated:* will trigger the Lambda function whenever a new object is created in the S3 bucket. Using a filter rule to generate notifications only when the suffix includes .csv will ensure that the Lambda function only runs for .csv files. Setting the ARN of the Lambda function as the destination for the event notification will directly invoke the Lambda function without any additional steps.

Option B is incorrect because it requires the user to tag the objects with .csv, which adds an extra step and increases the operational overhead.

Option C is incorrect because it uses an event type of s3:*, which will trigger the Lambda function for any S3 event, not just object creation. This could result in unnecessary invocations and increased costs.

Option D is incorrect because it involves creating and subscribing to an SNS topic, which adds an extra layer of complexity and operational overhead.

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide, Chapter 3: Data Ingestion and Transformation, Section 3.2: S3 Event Notifications and Lambda Functions, Pages 67-69

Building Batch Data Analytics Solutions on AWS, Module 4: Data Transformation, Lesson 4.2: AWS Lambda, Pages 4-8

AWS Documentation Overview, AWS Lambda Developer Guide, Working with AWS Lambda Functions, Configuring Function Triggers, Using AWS Lambda with Amazon S3, Pages 1-5

The correct answer is B because the workload requires complex processing that can run for 5 to 30 minutes, automatic scaling, and exactly-once-style avoidance of reprocessing source files. AWS Glue jobs are designed for serverless ETL processing over S3 data, and job bookmarks track previously processed source data so the job processes only new data on later runs. AWS documentation states that Glue job bookmarks help prevent reprocessing of old data and can track S3 source files for supported formats. Lambda is a poor fit because it has execution-duration limits and is not ideal for 30-minute complex transformations. EMR and EC2 polling add cluster and infrastructure management. Glue jobs with bookmarks provide the least operational overhead for this data engineering pattern.

Option C is the most cost-effective because it keeps the datasets in Amazon S3 and uses Amazon Athena to query them with SQL only when needed, avoiding the cost of running always-on database infrastructure. The study material states that “Amazon Athena is a serverless service that allows you to query data stored in Amazon S3 using standard SQL” . This directly matches the requirement that analysts will “analyze the data by using SQL queries,” while remaining cost-efficient due to serverless, on-demand querying.

To make CSV and JSON in S3 easily queryable, metadata must be discoverable and managed. The material also highlights that AWS Glue automates cataloging data in S3 through the AWS Glue Data Catalog, which “helps discover and manage metadata for data stored in AWS,” enabling query engines to treat file data as tables .

Other options are less aligned: RDS MySQL adds continuous capacity and administration costs; DataBrew is primarily for visual preparation (clean/normalize) rather than serving as the SQL query layer ; and QuickSight/SPICE targets BI dashboards, not general-purpose ad hoc SQL over raw files.

Option C is correct because the main performance problem here is the large number of very small files. AWS Glue documentation explains that for Apache Iceberg tables, managed compaction reduces metadata overhead and improves read performance by compacting many small objects into larger ones. AWS Glue also provides table optimizers for Iceberg, and compaction can be configured to run automatically. This directly addresses the requirement that performance should not degrade as more files are added over time.

Option A could help somewhat, but it adds more operational work and the crawler step is unnecessary noise for this question. With Iceberg tables in the Glue Data Catalog, the better answer is the native automatic compaction capability rather than a manually scheduled Glue job. Option B is incorrect because the Data Catalog does not simply compact files “every minute” by default; compaction is enabled and configured through the Iceberg table optimizer settings. Option D is insufficient because partitioning alone does not solve the small-files problem, and frequent crawler runs do not prevent long-term Iceberg degradation. The AWS-native, least-operations answer is to enable automatic compaction based on thresholds.

The correct solution is C: AWS Secrets Manager.

Why? AWS Secrets Manager is a fully managed service that helps you protect access to your applications, services, and IT resources without the upfront cost and complexity of managing your own hardware security module (HSM) infrastructure.

It allows for automatic rotation of secrets without requiring changes to the application code, meeting the requirement of minimal operational overhead.

Applications can securely retrieve credentials using Secrets Manager APIs, and the service integrates with AWS Identity and Access Management (IAM) to control access to secrets.

“You can use Secrets Manager to store credentials and to configure automatic rotation.”

In this scenario, the company is using Amazon Elastic Kubernetes Service (EKS) and wants secure access to DynamoDB without embedding credentials inside the application containers. The best practice is to use IAM roles for service accounts (IRSA), which allows assigning IAM roles to Kubernetes service accounts. This lets the EKS pods assume specific IAM roles securely, without the need to store credentials in containers.

IAM Roles for Service Accounts (IRSA):

With IRSA, each pod in the EKS cluster can assume an IAM role that grants access to DynamoDB without needing to manage long-term credentials. The IAM role can be attached to the service account associated with the pod.

This ensures least privilege access, improving security by preventing credentials from being embedded in the containers.

 AWS DataSync is an online data movement and discovery service that simplifies and accelerates data migrations to AWS as well as moving data to and from on-premises storage, edge locations, other cloud providers, and AWS Storage services1. AWS DataSync can copy data to and from various sources and targets, including Amazon S3, and handle files in multiple formats. AWS DataSync also supports incremental transfers, meaning it can detect and copy only the changes to the data, reducing the amount of data transferred and improving the performance. AWS DataSync can automate and schedule the transfer process using triggers, and monitor the progress and status of the transfers using CloudWatch metrics and events1.

AWS DataSync is the most operationally efficient way to transfer the data in this scenario, as it meets all the requirements and offers a serverless and scalable solution. AWS Glue, AWS Direct Connect, and Amazon S3 Transfer Acceleration are not the best options for this scenario, as they have some limitations or drawbacks compared to AWS DataSync. AWS Glue is a serverless ETL service that can extract, transform, and load data from various sources to various targets, including Amazon S32. However, AWS Glue is not designed for large-scale data transfers, as it has some quotas and limits on the number and size of files it can process3. AWS Glue also does not support incremental transfers, meaning it would have to copy the entire data set every time, which would be inefficient and costly.

AWS Direct Connect is a service that establishes a dedicated network connection between your on-premises data center and AWS, bypassing the public internet and improving the bandwidth and performance of the data transfer. However, AWS Direct Connect is not a data transfer service by itself, as it requires additional services or tools to copy the data, such as AWS DataSync, AWS Storage Gateway, or AWS CLI. AWS Direct Connect also has some hardware and location requirements, and charges you for the port hours and data transfer out of AWS.

Amazon S3 Transfer Acceleration is a feature that enables faster data transfers to Amazon S3 over long distances, using the AWS edge locations and optimized network paths. However, Amazon S3 Transfer Acceleration is not a data transfer service by itself, as it requires additional services or tools to copy the data, such as AWS CLI, AWS SDK, or third-party software. Amazon S3 Transfer Acceleration also charges you for the data transferred over the accelerated endpoints, and does not guarantee a performance improvement for every transfer, as it depends on various factors such as the network conditions, the distance, and the object size. References:

AWS DataSync

AWS Glue

AWS Glue quotas and limits

[AWS Direct Connect]

[Data transfer options for AWS Direct Connect]

[Amazon S3 Transfer Acceleration]

[Using Amazon S3 Transfer Acceleration]

Option A is correct because AWS Glue crawlers are the AWS-native service for discovering metadata and updating the AWS Glue Data Catalog. AWS documentation says that crawlers can crawl data stores, infer schema, and upon completion create or update tables in the Data Catalog. For schema-change tracking on an Amazon Redshift table, the crawler can connect through JDBC, inspect the table metadata, and update the Glue Data Catalog on a schedule. This directly satisfies the requirement to implement a cataloging solution that tracks schema changes over time.

Option B is incorrect because AWS DataSync is for data transfer, not metadata cataloging. Option C is also incorrect because AWS SCT is intended for schema conversion and migration assessments, not ongoing catalog synchronization into a Hive metastore for this use case. Option D is not the best answer because the requirement is for an AWS cataloging solution, and the standard managed catalog service here is the AWS Glue Data Catalog, not an external Apache Hive metastore. The daily scheduled crawler against Redshift with Glue Data Catalog updates is the most direct and exam-aligned solution for tracking schema changes.

The best solution for managing SSL/TLS certificates on EC2 instances with minimal operational overhead is to use AWS Certificate Manager (ACM). ACM simplifies certificate management by automating the provisioning, renewal, and deployment of certificates.

AWS Certificate Manager (ACM):

ACM manages SSL/TLS certificates for EC2 and other AWS resources, including automatic certificate renewal. This reduces the need for manual management and avoids operational complexity.

ACM also integrates with other AWS services to simplify secure connections between AWS infrastructure and customer-managed environments.

Option A is correct because AWS Glue is a serverless ETL service built to run PySpark workloads at scale with minimal infrastructure management. AWS documentation states that you can install additional Python modules and libraries for use with AWS Glue ETL jobs, including by using the --additional-python-modules parameter and Amazon S3 paths for wheel artifacts or other supported package delivery methods. That directly addresses the requirement for custom internal libraries. Since the data is already in Amazon S3 and the result must be loaded into Amazon Redshift, Glue is a natural low-overhead fit for this daily transformation pipeline.

Option B and C require managing compute infrastructure or cluster lifecycle, which increases operational overhead. Option D is not the best fit because SageMaker Processing is designed primarily for ML-oriented data preparation, not as the standard AWS service for large-scale ETL into Redshift. The question explicitly asks for least operational overhead with scalable PySpark and custom libraries, and AWS Glue provides exactly that managed capability.

 Using an S3 bucket that is in the same AWS Region where the company runs Athena queries can improve query performance by reducing data transfer latency and costs. Preprocessing the .csv data to Apache Parquet format can also improve query performance by enabling columnar storage, compression, and partitioning, which can reduce the amount of data scanned and fetched by the query. These solutions can optimize the storage solution for the POC test without requiring much effort or changes to the existing data pipeline. The other solutions are not optimal or relevant for this requirement. Adding a randomized string to the beginning of the keys in Amazon S3 can improve the throughput across partitions, but it can also make the data harder to query and manage. Using an S3 bucket that is in the same account that uses Athena to query the data does not have any significant impact on query performance, as long as the proper permissions are granted. Preprocessing the .csv data to JSON format does not offer any benefits over the .csv format, as both are row-based and verbose formats that require more data scanning and fetching than columnar formats like Parquet. References:

Best Practices When Using Athena with AWS Glue

Optimizing Amazon S3 Performance

AWS Certified Data Engineer - Associate DEA-C01 Complete Study Guide

Option A is the only choice that directly addresses the root cause: inconsistent postal-code formatting causing processing errors. In AWS Glue, a DynamicFrame can encounter issues when incoming data contains unexpected types or malformed values. Providing an explicit PySpark schema forces the postal code column to be interpreted consistently (for example, as a string with the expected structure), which prevents downstream steps from failing because of unexpected characters or mixed representations. After enforcing a consistent schema, the Glue job can standardize values (such as trimming extra digits, removing invalid characters, and normalizing case) and then write the corrected output back to Amazon S3 so the data lake is fixed at the source.

The document reinforces that AWS Glue is the managed ETL service used to transform data and move it between stores, which is exactly what is needed to correct bad values and persist clean outputs back into the lake. It also highlights that AWS Glue can perform validation as part of the ETL process, supporting the idea that data can be checked and corrected during processing rather than allowing bad values to break the workflow.

Options B, C, and D do not fix data quality or formatting issues; they relate to workflow state sharing, partition filtering, or internal implementation details, not cleansing invalid postal codes.