Event volume changes greatly from day to day and is difficult to estimate in advance.
Avoid manually changing shard counts while capacity follows incoming traffic.
Which stream setting is most appropriate?
Use Kinesis Data Streams on-demand mode to manage capacity automatically as traffic changes.
Detailed explanation
Lambda memory affects consumer compute capacity, not the stream's write capacity.
Lambda memory affects consumer compute capacity, not the stream's write capacity.
On-demand mode automatically manages capacity as traffic changes and reduces manual shard planning.
On-demand mode automatically manages capacity as traffic changes and reduces manual shard planning.
A fixed shard may not handle a peak and can cause capacity shortages or throttling.
A fixed shard may not handle a peak and can cause capacity shortages or throttling.
S3 versioning manages object versions and does not adjust stream ingestion capacity.
S3 versioning manages object versions and does not adjust stream ingestion capacity.
Check official documentation
Compare the following points with the official documentation.
Kinesis Data Streamsのオンデマンドモードとプロビジョンドモードの公式説明を確認する。
What to confirm
ストリーム容量の自動管理と、コンシューマーの計算リソースを区別できる。
Key points
- On-demand mode
- Provisioned mode
- Traffic variability
- Capacity management
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
Mode selection
Provisioned mode suits predictable traffic; on-demand mode reduces planning for variable workloads.
Cost and limits
Automatic scaling does not remove the need to review quotas, limits, and pricing.