Analytical queries and table sizes change over time, making a fixed distribution key hard to maintain.
Reduce manual tuning while validating the physical design recommended for the workload.
Which design is most appropriate?
Consider DISTSTYLE AUTO and validate its changes with Advisor, EXPLAIN plans, and redistribution metrics.
Detailed explanation
Replicating large tables increases storage, load, and maintenance costs.
Replicating large tables increases storage, load, and maintenance costs.
Even distribution avoids some skew but can increase redistribution for frequent joins.
Even distribution avoids some skew but can increase redistribution for frequent joins.
Externalizing data alone does not solve joins and query requirements inside Redshift.
Externalizing data alone does not solve joins and query requirements inside Redshift.
Redshift can select and adjust a distribution style based on table size and workload, but the effect must be measured.
Redshift can select and adjust a distribution style based on table size and workload, but the effect must be measured.
Check official documentation
Compare the following points with the official documentation.
Amazon Redshift DISTSTYLE AUTO、Automatic Table Optimization、Advisorの公式説明を確認する。
What to confirm
自動分散が無条件の最適解ではなく、ワークロード測定と併用することを説明できる。
Key points
- DISTSTYLE AUTO
- Automatic Table Optimization
- Advisor
- Redistribution
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
Automatic does not mean guaranteed
AUTO still needs observation of timing, plan changes, and data skew.
Evidence-based tuning
EXPLAIN plans and system metrics connect a physical change to an actual workload effect.