Understanding Statistics in SQL Server
summary
Explains how SQL Server uses column and index statistics to drive cardinality estimates, shows DBCC SHOW_STATISTICS output (header, density vector, histogram) and matches estimates to actual row counts in demos, and demonstrates that skewed distributions and sampling cause large plan differences: examples include a 5,295 correct estimate, a 5,437 near-miss, and queries returning 22k vs estimated 5k rows. Covers auto-create/update behavior, auto update statistics asynchronously, effects of updates on plan cache and parameter sniffing / parameter sensitivity, and using DBCC SHOW_STATISTICS to decide fixes.
AI-written summary · about
chapters (6)
- 0:00Start
- 4:23Statistics-related Database Options
- 7:02Query or join on a column, get a statistic!
- 25:33How do we manage statistics?
- 26:28Updating statistics
- 29:30Threshold Comparison
related talks
53mThe Most Important Performance Factor in SQL Server and Azure SQL- 53mThe Most Important Performance Factor in SQL Server and Azure SQL
52mA Deep Dive into Postgres Statistics
53mSQL Server Internals: The Practical Angle
04mEnhancing Planner Estimates with Extended Statistics in PostgreSQL
42mMariaDB and MySQL — What Statistics the Optimizer Uses
49mSQL Server Query Execution: Expectation vs Reality
53mIndexing Internals for Developers & DBAs