AWS Certified Machine Learning – Specialty — Question 283

A company plans to build a custom natural language processing (NLP) model to classify and prioritize user feedback. The company hosts the data and all machine learning (ML) infrastructure in the AWS Cloud. The ML team works from the company's office, which has an IPsec VPN connection to one VPC in the AWS Cloud.

The company has set both the enableDnsHostnames attribute and the enableDnsSupport attribute of the VPC to true. The company's DNS resolvers point to the VPC DNS. The company does not allow the ML team to access Amazon SageMaker notebooks through connections that use the public internet. The connection must stay within a private network and within the AWS internal network.

Which solution will meet these requirements with the LEAST development effort?

Answer options

• A. Create a VPC interface endpoint for the SageMaker notebook in the VPC. Access the notebook through a VPN connection and the VPC endpoint.
• B. Create a bastion host by using Amazon EC2 in a public subnet within the VPC. Log in to the bastion host through a VPN connection. Access the SageMaker notebook from the bastion host.
• C. Create a bastion host by using Amazon EC2 in a private subnet within the VPC with a NAT gateway. Log in to the bastion host through a VPN connection. Access the SageMaker notebook from the bastion host.
• D. Create a NAT gateway in the VPC. Access the SageMaker notebook HTTPS endpoint through a VPN connection and the NAT gateway.

Correct answer: A

Explanation

Creating an interface VPC endpoint (AWS PrivateLink) for Amazon SageMaker notebooks allows secure, private access directly over the IPsec VPN without traversing the public internet, requiring minimal configuration. Options B and C require provisioning and managing Amazon EC2 bastion hosts, which increases operational and development overhead. Option D is incorrect because a NAT gateway is designed for outbound internet access and does not establish a private connection to AWS services inside the internal network.