Data Warehousing on AWS Training Data Warehousing on AWS: Trends, Network And Critical Thinking 2 — Questions and Answers
Question 1: A company wants to isolate their Amazon Redshift cluster from the public internet. Which networking approach should they use?
- Deploy the cluster in a VPC with private subnets and no internet gateway route (Correct answer)
- Use Redshift's built-in firewall to block all external traffic
- Enable Redshift Enhanced VPC Routing and configure public subnets only
- Attach a NAT gateway directly to the Redshift cluster
Correct answer: Deploy the cluster in a VPC with private subnets and no internet gateway route
Deploying Redshift in a VPC with private subnets and no internet gateway route ensures the cluster is inaccessible from the public internet.
Question 2: Which Redshift feature forces all COPY and UNLOAD traffic to flow through the VPC rather than over the public internet?
- VPC Peering
- Enhanced VPC Routing (Correct answer)
- PrivateLink Endpoints
- VPC Flow Logs
Correct answer: Enhanced VPC Routing
Enhanced VPC Routing forces all data traffic between Redshift and Amazon S3 (or other services) to travel through the VPC, enabling VPC security controls.
Question 3: A data warehouse team notices their ETL jobs are slow due to network latency between their EC2 ETL servers and Redshift. What is the most effective network optimization?
- Place EC2 instances and Redshift in the same Availability Zone within the same VPC (Correct answer)
- Enable Redshift Spectrum to offload processing to S3
- Increase the Redshift node type to ra3.16xlarge
- Use AWS Direct Connect from the EC2 instances to Redshift
Correct answer: Place EC2 instances and Redshift in the same Availability Zone within the same VPC
Co-locating EC2 ETL servers and Redshift in the same AZ within the same VPC minimizes network latency by avoiding cross-AZ data transfer.
Question 4: Modern data warehousing on AWS increasingly separates compute from storage. Which Redshift node type is designed specifically to support this architecture?
- dc2.large
- ds2.xlarge
- ra3.xlplus (Correct answer)
- dc1.8xlarge
Correct answer: ra3.xlplus
RA3 nodes decouple compute and storage by using Redshift Managed Storage (RMS) backed by S3, allowing independent scaling of each layer.
Question 5: A company must ensure that data transferred between their on-premises systems and Amazon Redshift is encrypted in transit. Which approach satisfies this requirement?
- Enable SSL/TLS on the Redshift cluster JDBC/ODBC connection (Correct answer)
- Use AWS KMS to encrypt the Redshift cluster at rest
- Enable Redshift Enhanced VPC Routing
- Configure an S3 bucket policy requiring server-side encryption
Correct answer: Enable SSL/TLS on the Redshift cluster JDBC/ODBC connection
Enabling SSL/TLS on JDBC/ODBC connections encrypts data in transit between client applications and the Redshift cluster.
Question 6: Which emerging trend in data warehousing does Amazon Redshift Serverless primarily address?
- Eliminating the need for SQL-based analytics
- Removing the requirement to provision and manage cluster capacity (Correct answer)
- Replacing columnar storage with row-based storage for faster inserts
- Enabling real-time streaming without any batch processing
Correct answer: Removing the requirement to provision and manage cluster capacity
Redshift Serverless automatically provisions and scales capacity, removing the need for administrators to manually choose node types or cluster sizes.
Question 7: A security team requires that all access to Amazon Redshift must use AWS IAM authentication instead of database passwords. How is this achieved?
- Enable IAM database authentication in the Redshift cluster parameter group
- Use GetClusterCredentials API to generate temporary database credentials via IAM (Correct answer)
- Attach an IAM role directly to each Redshift database user
- Configure AWS Cognito as the identity provider for Redshift
Correct answer: Use GetClusterCredentials API to generate temporary database credentials via IAM
The GetClusterCredentials API allows IAM users and roles to obtain temporary database credentials, enabling IAM-based authentication without static passwords.
A company wants to isolate their Amazon Redshift cluster from the public internet.
Which networking approach should they use?