A Kinesis Data Streams consumer is falling behind and event delay is increasing.
Choose evidence for shard scaling or processing improvements.
Which monitoring approach is most appropriate?
Track IteratorAgeMilliseconds together with throughput and throttling metrics to distinguish input, shard, and consumer bottlenecks.
Detailed explanation
Iterator age shows how old the oldest unprocessed record is and exposes consumer lag.
Iterator age shows how old the oldest unprocessed record is and exposes consumer lag.
S3 retention is unrelated to Kinesis consumer lag.
S3 retention is unrelated to Kinesis consumer lag.
Producer API volume does not measure consumer delay.
Producer API volume does not measure consumer delay.
A total does not show the difference between ingest rate and processing rate.
A total does not show the difference between ingest rate and processing rate.
Check official documentation
Compare the following points with the official documentation.
Amazon Kinesis Data Streams IteratorAgeMilliseconds、CloudWatch metrics、shard scalingの公式説明を確認する。
What to confirm
投入量と処理遅延を別メトリクスで見る理由を説明できる。
Key points
- IteratorAge
- Shard
- Throughput
- Backlog
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
Lag decomposition
Separate producer rate, shard capacity, consumer runtime, and checkpoint behavior.
Scaling choice
Compare on-demand, resharding, and enhanced fan-out cost with the lag reduction.