Retrieve customers, orders, and order history at low latency from customer and order identifiers.
Define access patterns first and avoid a relational JOIN for the main reads.
Which data-store design is most appropriate?
Use a DynamoDB single-table design with composite keys and item collections that match the required queries.
Detailed explanation
PK and SK can colocate related entities so a defined access pattern is served by a Query.
PK and SK can colocate related entities so a defined access pattern is served by a Query.
An object key does not provide conditional updates, low-latency Query, or concurrency control.
An object key does not provide conditional updates, low-latency Query, or concurrency control.
Athena is for analytical scans and is not a low-latency single-customer lookup.
Athena is for analytical scans and is not a low-latency single-customer lookup.
Relational joins are flexible but may not meet the required millisecond key access and scale.
Relational joins are flexible but may not meet the required millisecond key access and scale.
Check official documentation
Compare the following points with the official documentation.
DynamoDB single-table design、Query、partition key・sort keyの公式説明を確認する。
What to confirm
RDSの柔軟なJOINとDynamoDBのアクセスパターン設計を比較できる。
Key points
- Single-table design
- Partition key
- Sort key
- Access pattern
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
Access-pattern-first modeling
Define reads and writes before choosing the PK and SK.
Controlled denormalization
Duplicate selected attributes when it reduces joins while keeping updates consistent.