A company hosts a data ingestion service on Amazon EC2 instances in a private subnet. The service writes of data monthly to Amazon DynamoDB and sends of message payloads monthly to Amazon Simple Queue Service (Amazon SQS). Currently, all outbound traffic from the private subnet is routed through a NAT Gateway. A solutions architect must design a routing solution to minimize the data transfer and processing costs associated with these services. Which architecture meets these requirements most cost-effectively?
- ACreate Gateway VPC endpoints for both Amazon DynamoDB and Amazon SQS. Configure the route table for the private subnet to route all traffic to both services through the gateway endpoints.
- Create a Gateway VPC endpoint for Amazon DynamoDB and an Interface VPC endpoint (AWS PrivateLink) for Amazon SQS. Configure the route table for the private subnet to route DynamoDB traffic through the gateway endpoint.Cevap
- CCreate Interface VPC endpoints (AWS PrivateLink) for both Amazon DynamoDB and Amazon SQS. Configure the route table for the private subnet to route all traffic to both services through the interface endpoints.
- DKeep routing all traffic through the NAT Gateway, but purchase a Compute Savings Plan to offset the NAT Gateway data processing charges.
Cevap
Creating a Gateway VPC endpoint for Amazon DynamoDB and an Interface VPC endpoint (AWS PrivateLink) for Amazon SQS, and routing DynamoDB traffic through the gateway endpoint.
The architecture utilizing a Gateway VPC endpoint for Amazon DynamoDB and an Interface VPC endpoint for Amazon SQS is the most cost-effective. Gateway VPC endpoints are free of charge, making them ideal for high-volume data transfers (/month). Amazon SQS does not support Gateway VPC endpoints, so an Interface VPC endpoint (AWS PrivateLink) must be used. Although Interface VPC endpoints incur hourly and per-gigabyte processing fees, using one for the low-volume SQS traffic (/month) avoids the higher NAT Gateway processing fee () and keeps the traffic private.
Adım Adım Çözüm
Anahtar Kavram
VPC Endpoint cost optimization and compatibility