1. Which of the following Amazon Elastic MapReduce (EMR) cluster configurations would be most suitable for a machine learning (ML) workload that requires high memory resources and distributed processing across a large number of instances?

A) A cluster with all instances running on r5.8xlarge with EBS-optimized instances and a single master node running on an m5.8xlarge instance.
B) A cluster with a mix of instance types, such as c5.xlarge and r5.2xlarge, with EBS-optimized instances and a single master node running on an m5.4xlarge instance.
C) A cluster with all instances running on r5.4xlarge with EBS-optimized instances and a single master node running on an m5.4xlarge instance.
D) A cluster with all instances running on r5d.4xlarge with EBS-optimized instances and a single master node running on an m5.2xlarge instance.



2. A retail company wants to collect customer transaction data from multiple sources, such as point-of-sale systems and online purchases, for analysis. Which AWS service is best suited for this use case?

A) Amazon S3
B) Amazon Kinesis Data Streams
C) Amazon QuickSight
D) AWS Glue
E) Amazon Aurora


3. Which of the following statements is true about using Amazon Elastic Kubernetes Service (EKS) for deploying machine learning models?

A) EKS can be used for deploying machine learning models with CPU instances, but GPU instances are not supported.
B) EKS is not suitable for machine learning workloads because it lacks GPU support.
C) EKS supports GPU instances, which are ideal for training and deploying machine learning models.
D) EKS does not support deploying machine learning models that require more than 16 GB of memory.



4. Which of the following statements is true regarding Amazon Translate and its use in machine learning applications?

A) Amazon Translate is a supervised learning algorithm that can be used to train custom machine translation models for specific domains.
B) Amazon Translate is a rule-based machine translation service that requires extensive training data to provide accurate translations.
C) Amazon Translate supports automatic language detection, allowing you to translate text without having to specify the source language.
D) Amazon Translate is a cloud-based neural machine translation service that provides real-time translation of text from one language to another.
E) Amazon Translate can be used to translate large volumes of text quickly and accurately, but it does not provide customizable translation models for specific domains.


5. A team of developers is building a machine learning application and wants to store Docker images containing the application code and dependencies. The application uses GPU-based machine learning models and requires a custom deep learning framework. The team has decided to use Amazon Elastic Container Registry (Amazon ECR) to store their Docker images. Which of the following statements is true regarding the use of Amazon ECR for this purpose?

A) Amazon ECR is not suitable for storing GPU-based workloads or custom deep learning frameworks, as it only supports standard Docker images.
B) Amazon ECR is a fully managed container registry service that allows users to store, manage, and deploy Docker images. It supports GPU-based workloads and custom deep learning frameworks, and provides integrations with Amazon ECS, Amazon EKS, and AWS Batch.
C) Amazon ECR is only suitable for storing CPU-based workloads, as GPU-based workloads require more complex infrastructure and are better suited to Amazon SageMaker.
D) Amazon ECR can store GPU-based workloads and custom deep learning frameworks, but it requires additional configuration and management to set up the necessary infrastructure.
E) Amazon ECR is not suitable for machine learning applications, as it is designed primarily for storing standard Docker images used in web applications and microservices.