- (Exam Topic 2)
A company has loT sensors that monitor traffic patterns throughout a large city. The company wants to read and collect data from the sensors and perform aggregations on the data.
A solutions architect designs a solution in which the loT devices are streaming to Amazon Kinesis Data Streams. Several applications are reading from the stream. However, several consumers are experiencing throttling and are periodically and are periodically encountering a RealProvisioned Throughput Exceeded error.
Which actions should the solution architect take to resolve this issue? (Select THREE.)

1. A. Reshard the stream to increase the number of shards s in the stream.
2. B. Use the Kinesis Producer Library KPL). Adjust the polling frequency.
3. C. Use consumers with the enhanced fan-out feature.
4. D. Reshard the stream to reduce the number of shards in the stream.
5. E. Use an error retry and exponential backoff mechanism in the consumer logic.
6. F. Configure the stream to use dynamic partitioning.

Correct Answer: ACE
https://repost.aws/knowledge-center/kinesis-readprovisionedthroughputexceeded Follow Data Streams best practices
To mitigate ReadProvisionedThroughputExceeded exceptions, apply these best practices:
• Reshard your stream to increase the number of shards in the stream.
• Use consumers with enhanced fan-out. For more information about enhanced fan-out, see Developing custom consumers with dedicated throughput (enhanced fan-out).
• Use an error retry and exponential backoff mechanism in the consumer logic if ReadProvisionedThroughputExceeded exceptions are encountered. For consumer applications that use an AWS SDK, the requests are retried by default.

- (Exam Topic 3)
A company is running a serverless application that consists of several AWS Lambda functions and Amazon DynamoDB tables. The company has created new functionality that requires the Lambda functions to access an Amazon Neptune DB cluster. The Neptune DB cluster is located in three subnets in a VPC.
Which of the possible solutions will allow the Lambda functions to access the Neptune DB cluster and DynamoDB tables? (Select TWO.)

1. A. Create three public subnets in the Neptune VPC, and route traffic through an internet gatewa
2. B. Host the Lambda functions in the three new public subnets.
3. C. Create three private subnets in the Neptune VPC, and route internet traffic through a NAT gateway.Host the Lambda functions in the three new private subnets.
4. D. Host the Lambda functions outside the VP
5. E. Update the Neptune security group to allow access from the IP ranges of the Lambda functions.
6. F. Host the Lambda functions outside the VP
7. G. Create a VPC endpoint for the Neptune database, and have the Lambda functions access Neptune over the VPC endpoint.
8. H. Create three private subnets in the Neptune VP
9. I. Host the Lambda functions in the three new isolatedsubnet
10. J. Create a VPC endpoint for DynamoDB, and route DynamoDB traffic to the VPC endpoint.

Correct Answer: BE
This option allows the company to use private subnets and VPC endpoints to connect the Lambda functions to the Neptune DB cluster and DynamoDB tables securely and efficiently1. By creating three private subnets in the Neptune VPC, the company can isolate the Lambda functions from the public internet and reduce the attack surface2. By routing internet traffic through a NAT gateway, the company can enable the Lambda functions to access AWS services that are outside the VPC, such as Amazon S3 or Amazon
CloudWatch3. By hosting the Lambda functions in the three new private subnets, the company can ensure that the Lambda functions can access the Neptune DB cluster within the same VPC4. By creating a VPC endpoint for DynamoDB, the company can enable the Lambda functions to access DynamoDB tables without going through the internet or a NAT gateway5. By routing DynamoDB traffic to the VPC endpoint, the company can improve the performance and availability of the DynamoDB access5.
References:
Configuring a Lambda function to access resources in a VPC
Working with VPCs and subnets
NAT gateways
Accessing Amazon Neptune from AWS Lambda
VPC endpoints for DynamoDB

- (Exam Topic 2)
A solutions architect needs to review the design of an Amazon EMR cluster that is using the EMR File System (EMRFS). The cluster performs tasks that are critical to business needs. The cluster is running Amazon EC2 On-Demand Instances at all times tor all task, primary, and core nodes. The EMR tasks run each morning, starting at 1 ;00 AM. and take 6 hours to finish running. The amount of time to complete the processing is not a priority because the data is not referenced until late in the day.
The solutions architect must review the architecture and suggest a solution to minimize the compute costs. Which solution should the solutions architect recommend to meet these requirements?

1. A. Launch all task, primary, and core nodes on Spool Instances in an instance flee
2. B. Terminate the cluster, including all instances, when the processing is completed.
3. C. Launch the primary and core nodes on On-Demand Instance
4. D. Launch the task nodes on Spot Instances in an instance flee
5. E. Terminate the cluster, including all instances, when the processing is complete
6. F. Purchase Compute Savings Plans to cover the On-Demand Instance usage.
7. G. Continue to launch all nodes on On-Demand Instance
8. H. Terminate the cluster, including all instances, when the processing is complete
9. I. Purchase Compute Savings Plans to cover the On-Demand Instance usage
10. J. Launch the primary and core nodes on On-Demand Instance
11. K. Launch the task nodes on Spot Instances in an instance flee
12. L. Terminate only the task node instances when the processing is complete
13. M. Purchase Compute Savings Plans to cover the On-Demand Instance usage.

Correct Answer: A
Amazon EC2 Spot Instances offer spare compute capacity at steep discounts compared to On-Demand prices. Spot Instances can be interrupted by EC2 with two minutes of notification when EC2 needs the capacity back. Amazon EMR can handle Spot interruptions gracefully by decommissioning the nodes and redistributing the tasks to other nodes. By launching all nodes on Spot Instances in an instance fleet, the solutions architect can minimize the compute costs of the EMR cluster. An instance fleet is a collection of EC2 instances with different types and sizes that EMR automatically provisions to meet a defined target capacity. By terminating the cluster when the processing is completed, the solutions architect can avoid paying for idle resources. References:
https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-managed-scaling.html
https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-instance-fleet.html
https://aws.amazon.com/blogs/big-data/optimizing-amazon-emr-for-resilience-and-cost-with-capacity-opt

- (Exam Topic 1)
A company has migrated its forms-processing application to AWS. When users interact with the application, they upload scanned forms as files through a web application. A database stores user metadata and references to files that are stored in Amazon S3. The web application runs on Amazon EC2 instances and an Amazon RDS for PostgreSQL database.
When forms are uploaded, the application sends notifications to a team through Amazon Simple Notification Service (Amazon SNS). A team member then logs in and processes each form. The team member performs data validation on the form and extracts relevant data before entering the information into another system that uses an API.
A solutions architect needs to automate the manual processing of the forms. The solution must provide accurate form extraction, minimize time to market, and minimize long-term operational overhead.
Which solution will meet these requirements?

1. A. Develop custom libraries to perform optical character recognition (OCR) on the form
2. B. Deploy the libraries to an Amazon Elastic Kubernetes Service (Amazon EKS) cluster as an application tie
3. C. Use this tier to process the forms when forms are uploade
4. D. Store the output in Amazon S3. Parse this output by extracting the data into an Amazon DynamoDB tabl
5. E. Submit the data to the target system's AP
6. F. Host the new application tier on EC2 instances.
7. G. Extend the system with an application tier that uses AWS Step Functions and AWS Lambd
8. H. Configure this tier to use artificial intelligence and machine learning (AI/ML) models that are trained and hosted on an EC2 instance to perform optical character recognition (OCR) on the forms when forms are uploade
9. I. Store the output in Amazon S3. Parse this output by extracting the data that is required within the application tie
10. J. Submit the data to the target system's API.
11. K. Host a new application tier on EC2 instance
12. L. Use this tier to call endpoints that host artificial intelligence and machine learning (Al/ML) models that are trained and hosted in Amazon SageMaker to perform optical character recognition (OCR) on the form
13. M. Store the output in Amazon ElastiCach
14. N. Parse this output by extracting the data that is required within the application tie
15. O. Submit the data to the target system's API.
16. P. Extend the system with an application tier that uses AWS Step Functions and AWS Lambd
17. Q. Configure this tier to use Amazon Textract and Amazon Comprehend to perform optical character recognition (OCR) on the forms when forms are uploade
18. R. Store the output in Amazon S3. Parse this output by extracting the data that is required within the application tie
19. S. Submit the data to the target system's API.

Correct Answer: D
Extend the system with an application tier that uses AWS Step Functions and AWS Lambda. Configure this tier to use Amazon Textract and Amazon Comprehend to perform optical character recognition (OCR) on the forms when forms are uploaded. Store the output in Amazon S3. Parse this output by extracting the data that is required within the application tier. Submit the data to the target system's API. This solution meets the requirements of accurate form extraction, minimal time to market, and minimal long-term operational overhead. Amazon Textract and Amazon Comprehend are fully managed and serverless services that can perform OCR and extract relevant data from the forms, which eliminates the need to develop custom libraries or train and host models. Using AWS Step Functions and Lambda allows for easy automation of the process and the ability to scale as needed.

- (Exam Topic 3)
A company is using AWS CodePipeline for the CI/CD of an application to an Amazon EC2 Auto Scaling group. All AWS resources are defined in AWS CloudFormation templates. The application artifacts are stored in an Amazon S3 bucket and deployed to the Auto Scaling group using instance user data scripts.
As the application has become more complex, recent resource changes in the CloudFormation templates have caused unplanned downtime.
How should a solutions architect improve the CI/CD pipeline to reduce the likelihood that changes in the templates will cause downtime?

1. A. Adapt the deployment scripts to detect and report CloudFormation error conditions when performing deployment
2. B. Write test plans for a testing team to execute in a non-production environment before approving the change for production.
3. C. Implement automated testing using AWS CodeBuild in a test environmen
4. D. Use CloudFormation change sets to evaluate changes before deploymen
5. E. Use AWS CodeDeploy to leverage blue/green deployment patterns to allow evaluations and the ability to revert changes, if needed.
6. F. Use plugins for the integrated development environment (IDE) to check the templates for errors, and usethe AWS CLI to validate that the templates are correc
7. G. Adapt the deployment code to check for error conditions and generate notifications on error
8. H. Deploy to a test environment and execute a manual test plan before approving the change for production.
9. I. Use AWS CodeDeploy and a blue/green deployment pattern with CloudFormation to replace the user data deployment script
10. J. Have the operators log in to running instances and go through a manual test plan to verify the application is running as expected.

Correct Answer: B