1. You need to deploy a machine learning model for real-time inference on AWS that requires the processing of large input data in batches. Which of the following AWS services would be the best fit for this use case?

A) AWS Lambda
B) Amazon SageMaker Neo
C) Amazon SageMaker Endpoints
D) Amazon EC2 Auto Scaling



2. You need to ingest and store streaming data from a fleet of IoT devices. Which AWS service would you use to ensure secure, scalable, and low-latency data ingestion and storage?

A) AWS IoT Core
B) Amazon S3
C) Amazon Elasticsearch Service
D) AWS Glue



3. Which of the following is a common preprocessing technique used in natural language processing (NLP) to normalize text before training a machine learning model?

A) None of the above
B) Mean Shift Clustering
C) Stop-word removal
D) Singular Value Decomposition (SVD)
E) Principal Component Analysis (PCA)


4. Which of the following is NOT a way to achieve high availability for Amazon Elastic Kubernetes Service (Amazon EKS)?

A) Using Amazon EKS managed node groups in different Availability Zones
B) Using multiple Availability Zones for the EKS control plane
C) Running EKS worker nodes on a single EC2 instance type and size
D) Configuring Kubernetes cluster autoscaling
E) Leveraging AWS Fargate as an alternative to running EKS worker nodes on EC2 instances


5. Which of the following statements is true regarding the use of Amazon Elastic File System (Amazon EFS) with Amazon SageMaker for machine learning workloads?

A) Amazon EFS is not supported for use with SageMaker's shared file system feature, as it only supports Amazon S3 as a data source.
B) Amazon EFS can be used as a data source for SageMaker, but it requires manual configuration of NFS mount points on the SageMaker instance.
C) Amazon EFS can be used as a highly scalable and highly available file system to store large datasets and intermediate results in SageMaker's shared file system feature.
D) Amazon EFS can only be used for storing small datasets and intermediate results in SageMaker's shared file system feature due to limitations in its scalability and availability.