A public REST API is receiving a sudden request spike that overloads the backend.
Control usage per consumer and limit excessive calls at the API entry point.
Which design is most appropriate?
Use an API Gateway usage plan with API keys and throttling to control request volume for each consumer.
Detailed explanation
API Gateway can associate API keys with usage plans and throttle calls for each consumer.
API Gateway can associate API keys with usage plans and throttle calls for each consumer.
KMS key management does not provide API request throttling.
KMS key management does not provide API request throttling.
Lifecycle manages S3 objects and does not rate-limit API calls.
Lifecycle manages S3 objects and does not rate-limit API calls.
DNS cache duration does not control the number of API calls reaching the backend.
DNS cache duration does not control the number of API calls reaching the backend.
Check official documentation
Compare the following points with the official documentation.
Amazon API Gatewayの使用量プランとスロットリングの公式説明を確認する。
What to confirm
APIの入口で利用量を制御し、バックエンドを過負荷から保護できる。
Key points
- Amazon API Gateway
- Usage plan
- Throttling
Notes
- Environment: AWS公式API Gatewayドキュメントの確認
- Confirm current AWS service behavior in the official documentation.
Foundation review
Boundary placement
Throttling at the API boundary protects the backend before requests reach application code.
Consumer isolation
Per-consumer quotas prevent one caller from consuming all available capacity.