Three applications read the same Kinesis Data Stream for detection, aggregation, and auditing.
Reduce read contention and keep latency low for each application.
Which feature is most appropriate?
Use Enhanced Fan-Out for each consumer so each receives dedicated read throughput.
Detailed explanation
Removing retained data does not separate consumer bandwidth and prevents replay.
Removing retained data does not separate consumer bandwidth and prevents replay.
Enhanced Fan-Out provides consumer-specific read throughput and a push-based delivery path.
Enhanced Fan-Out provides consumer-specific read throughput and a push-based delivery path.
Consumers share shard read capacity and can contend as their number increases.
Consumers share shard read capacity and can contend as their number increases.
This adds a storage and processing delay and does not provide low-latency stream delivery.
This adds a storage and processing delay and does not provide low-latency stream delivery.
Check official documentation
Compare the following points with the official documentation.
Kinesis Data StreamsのEnhanced Fan-Outとコンシューマー登録の公式説明を確認する。
What to confirm
共有読み取りとコンシューマー専用読み取りの遅延・スループット特性を説明できる。
Key points
- Enhanced Fan-Out
- Dedicated read throughput
- Multiple consumers
- Low latency
Notes
- Environment: AWS公式DEA-C01試験ガイドと各サービスの公式ドキュメントの確認
- Confirm current AWS service behavior in the official DEA-C01 documentation.
Foundation review
Shared versus dedicated reads
Shared GetRecords polling can contend for shard capacity; Enhanced Fan-Out separates consumer delivery.
Cost and capacity
Dedicated consumers improve isolation but require capacity and cost planning for registrations.