28.01.2026 –, 155
Jazyk: English
This talk introduces the key access paths for B-tree indexes in PostgreSQL. Based on over ten years of experience with Oracle’s index skip scan, I will highlight common problems it can cause and show practical ways to avoid them in PostgreSQL. I will briefly demo DB Booster, a tool that automatically finds missing indexes, and discuss the future of index tuning in relational databases—self-tuning engines and PostgreSQL’s place in that shift.
Whatever will fit in 45 minutes:
-Index tuning introduction - fast, effective, impactful
-Table without any index
-B-Tree: the basic index
-Why does the B-Tree index work
-Index Seek
-Index Scan
-Index Skip Scan
-Real-life issues with the Skip Scan in Oracle
-How to avoid these issues in PotsgreSQL
-Automated indexing - DB Booster
-The future of the databases - self-tuning engines
At sixteen, I fell in love with databases — first playing around with SQL Server 6.5 and 7.3, and after studying at the Faculty of Electrical Engineering (FEL, Czech Technical University), I fell for Oracle. PostgreSQL has stuck with me continuously throughout my career. I've always been curious about how databases work under the hood. I'm a developer by background, focused mainly on performance tuning. My long-term goal is to build a self-tuning PostgreSQL, at least as a PoC :)