Optimize your workload with automatic indexing
summary
Automatic indexing in Oracle Database 19c (with enhancements toward 23c) captures real workload via the automatic SQL tuning set, generates candidate indexes and builds them as invisible indexes for background test runs, then makes only those that measurably improve query performance visible. It now considers range predicates and accounts for DML overhead, supports function-based indexes, offers table-level include/exclude controls, performs non-blocking reparses, and can drop unused or even redundant manual indexes; the decision is always driven by measured workload benefit, not cost estimates.
AI-written summary · about
chapters (9)
- 0:00Start
- 2:35Indexes- pros and cons
- 7:39Index changes and the application lifecycle
- 9:15A traditional approach is complex
- 10:46Automatic indexing methodology
- 15:36How it works
- 18:29Reporting
- 19:01Enhancements since Oracle Database 19c
- 24:18Summary
related talks
20mImprove Your Data Strategy with Database 23ai
12mKey AI Innovations in Oracle Database 23ai
29mDo nothing and improve Autonomous Database performance
27mAutomating Data Management for All
24mFuture of Data and AI
01h 00m7 Tips to Performance Tuning, Optimization, and Everything
18mData Superpowers with Oracle Database 23ai
51mHow Modern SQL Databases Come up with Algorithms that You Would Have Never Dreamed Of