A large number of small sensor events must be sent to Kinesis Data Streams.
Reduce API calls and per-record overhead while consumers can still recover the original user records.
Which feature is most appropriate?
Use record aggregation in the Kinesis Producer Library (KPL) to combine user records into Kinesis records.
Detailed explanation
This obscures record boundaries, complicates retries, and can hit record-size limits.
This obscures record boundaries, complicates retries, and can hit record-size limits.
Deep Archive is for long-term storage and does not provide low-latency ingestion into Kinesis.
Deep Archive is for long-term storage and does not provide low-latency ingestion into Kinesis.
KPL combines multiple user records into one Kinesis record, improving sending efficiency while consumers can deaggregate them.
KPL combines multiple user records into one Kinesis record, improving sending efficiency while consumers can deaggregate them.
Log retention does not change Kinesis API call count or record aggregation.
Log retention does not change Kinesis API call count or record aggregation.
Check official documentation
Compare the following points with the official documentation.
Kinesis Producer Libraryのレコード集約、ユーザーレコード、集約解除の公式説明を確認する。
What to confirm
ストリームの保持設定と、送信側のレコード集約を区別できる。
Key points
- KPL
- Record aggregation
- User record
- API efficiency
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
Capacity units
Kinesis capacity and request limits differ between service records and aggregated user records.
Consumer compatibility
Consumers must understand the aggregation format and deaggregate records correctly.