AWR reporting. Enchance your db development and support
summary
AWR is a historical Oracle diagnostic store that takes snapshots hourly by default (configurable) and captures SQL text, plans, DBA_HIST views (162 in Oracle 23), and ASH samples (one per second retained ~30 minutes). Key actionable metric is active sessions ≈ DB_TIME / ELAPSED_TIME compared to CPU count to decide if the database is the bottleneck. Practical tools shown: scripts to export/import AWR, HTML/text reports, and a SQL Developer user report that aggregates many hourly AWRs — built and demonstrated working end to end.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 1:28What AWR is and why it matters
- 3:40AWR data collection
- 5:25Key AWR report data
- 7:55AWR data and docs
- 10:04Execution plans history
- 11:51Generating AWR reports
- 13:16AWR load profile
- 16:30AWR sections for devs
- 20:06Support troubleshooting challenges
- 22:54Case study: tracing vs AWR
- 25:25Custom AWR snapshots
- 26:52Reading AWR reports
- 28:41Wait events explained
- 30:07ASH metric significance
- 31:56When to use AWR
- 34:53CPU and workload bottlenecks
- 36:19Automating AWR analytics
- 39:09MODEL SQL for Delta
- 41:34Q&A
related talks
30mMoving your application to Oracle Autonomous Database
30mExadata warehouse: simplify capacity planning and forecasting for Exadata
34mOracle Autonomous Database: simplify your database management
54mSoaring to New Heights with Accelerated Database Recovery
28mNext-Gen, data-driven application development
30mSelf-service data warehousing
27mAutomating Data Management for All
28mSimplify data analysis using analytic views