An aggregation query has become much slower in Redshift.
Determine whether distribution, statistics, scanning, or the join plan caused the regression.
Which approach is most appropriate?
Compare EXPLAIN plans with execution metrics, waits, and statistics before changing distribution or cluster capacity.
Detailed explanation
The plan and actual row movement or scan work identify the likely physical or SQL cause.
The plan and actual row movement or scan work identify the likely physical or SQL cause.
Capacity may mask the symptom while leaving the underlying regression and cost unresolved.
Capacity may mask the symptom while leaving the underlying regression and cost unresolved.
API-event volume does not describe Redshift scans or joins.
API-event volume does not describe Redshift scans or joins.
S3 retention rules do not repair a Redshift query plan.
S3 retention rules do not repair a Redshift query plan.
Check official documentation
Compare the following points with the official documentation.
Amazon Redshift EXPLAIN、query monitoring rules、statisticsの公式説明を確認する。
What to confirm
無条件のスケールアップと計画に基づく性能分析の違いを説明できる。
Key points
- EXPLAIN
- DS_DIST
- Statistics
- WLM
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
Estimated versus actual
A large row-count difference can indicate stale statistics or skew.
Layered tuning
Evaluate SQL, physical layout, statistics, and capacity in sequence.