Events arriving in Firehose must be stored under an S3 prefix for each tenantId.
Choose the prefix from record attributes without creating a separate Lambda for every destination.
Which feature is most appropriate?
Use Firehose dynamic partitioning to calculate S3 prefixes from record attributes.
Detailed explanation
DMS mappings select or transform migration data and are not a Firehose delivery-prefix feature.
DMS mappings select or transform migration data and are not a Firehose delivery-prefix feature.
A consumer name identifies a stream reader and does not determine Firehose S3 prefixes.
A consumer name identifies a stream reader and does not determine Firehose S3 prefixes.
Website hosting serves objects and does not partition Firehose records by an attribute.
Website hosting serves objects and does not partition Firehose records by an attribute.
Dynamic partitioning uses record attributes to determine S3 prefixes and split delivered objects.
Dynamic partitioning uses record attributes to determine S3 prefixes and split delivered objects.
Check official documentation
Compare the following points with the official documentation.
Amazon Data Firehoseの動的パーティショニングとS3プレフィックス式の公式説明を確認する。
What to confirm
ストリームの読み取り設定と、配信時のS3パーティショニングを区別できる。
Key points
- Dynamic partitioning
- S3 prefix
- Record attribute
- Firehose
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
Partitioning trade-off
More dynamic keys can create smaller buffers and many small objects.
Delivery boundary
Dynamic partitioning occurs during delivery and differs from a stream consumer's partition assignment.