High Performance SQL with PostgreSQL
Deep-dive course by the recognised PostgreSQL contributing sponsor, Data Egret, designed to help developers build fast and reliable applications with PostgreSQL at scale.
Upcoming dates
Statistics for the Planner — 4 November 2026
10:00–12:30 Berlin time
Introduction to SQL Optimisations — 13 November 2026
10:00–11:30 Berlin time
SQL — What Not to Do — 25 November 2026
10:00–13:00 Berlin time
Best Practices for DDL/DML Migration — 7 December 2026
10:00–11:40 Berlin time
Understanding EXPLAIN — 17 December 2026
10:00–13:30 Berlin time
What you'll learn
The course covers advanced indexing techniques, query optimisation, SQL best practices, interpreting EXPLAIN plans, and more.
The course covers
- Advanced indexing techniques
- Query optimisation
- SQL best practices
- Interpreting EXPLAIN plans
Prerequisites
Working SQL: joins, aggregates, subqueries. No DBA experience and no PostgreSQL internals assumed.
For details see the course terms & conditions.
Syllabus
1Introduction to Execution Plans
This chapter introduces how PostgreSQL decides the best way to execute SQL queries. Key concepts include:
- Query planning and optimization: the optimizer evaluates multiple execution paths and chooses the most efficient based on cost estimation.
- Execution nodes: each plan is a tree of nodes like Seq Scan, Index Scan, or Join.
- Cost computation: the planner uses statistics about tables and indexes to estimate the execution cost.
- Tools for visualization: some online visualizers (e.g., explain.depesz.com ) help interpret query plans.
2Indexing Techniques
This chapter explains index structures and how to use them effectively:
- Types of indexes: introduction to PostgreSQL-supported index types (B-tree, GIN, GiST, BRIN, and others).
- Partial and functional indexes: these target specific use cases and optimize performance.
- Common pitfalls: some queries may not use indexes due to function usage, non-selective patterns (e.g., LIKE '%...' ), or invalid indexes.
- Best practices: choose indexes based on query patterns and ensure statistics are up to date.
3Statistics for the Planner
This chapter explores the role of statistics in query optimization:
- Types of statistics and their collection: managed by ANALYZE , which can be configured for frequency and detail.
- Challenges: stale or misleading stats can lead to poor plan choices.
- Extended stats: from version 10+, PostgreSQL can store correlations and dependencies across columns.
4SQL - What not to do
This chapter lists anti-patterns and common mistakes in SQL development:
- Bad modeling practices: avoid key-value schemas, multi-column fields, overly wide tables, or non-normalized designs.
- Misuse of SQL features: improper use of NULLs , implicit column orders, and excessive procedural logic in queries.
- Atomicity and constraints: always enforce data integrity with constraints.
- Inefficient queries: avoid overusing subqueries, and be mindful of indexing and pagination practices.
- Transaction management: manage locking, deadlocks, and transaction length to avoid contention.
5Introduction to SQL Optimisations
This chapter gives the first advice and best practices on writing performant SQL:
- Optimization targets: identify which queries to optimize through analysis.
- Query simplification: avoid unnecessary subqueries, rewrite complex joins, and remove unused views.
- Data access: limit the amount of data retrieved, avoid application-side filtering, and reduce connections overhead.
- Transaction management: manage locking, deadlocks, and transaction length to avoid contention.
- Scalability: introduction to data archiving and retention, best practices for large datasets including partitioning.
6Understanding EXPLAIN
This chapter explains how to read and interpret PostgreSQL’s execution plans:
- EXPLAIN options: EXPLAIN , EXPLAIN ANALYZE , with options like BUFFERS , WAL , and SETTINGS .
- Reading plans: focus on estimated vs actual rows, cost metrics, and node types.
- Common nodes: Seq Scan , Index Scan , Hash Join , etc.
- Troubleshooting: look for mismatched estimates, redundant joins, and costly sorts or writes.
Frequently asked questions
What prerequisites do I need?
Working SQL — joins, aggregates, subqueries — and comfort reading application code. No DBA experience and no PostgreSQL internals are assumed; this is a developer's course.
Can I take single sessions instead of the whole track?
Yes. Every session stands on its own and can be booked individually (€150 or €200 by length). Booking the full track costs €850 — €400 less than the seven sessions booked separately.
Will I receive a certificate?
Yes. Each session earns a certificate of attendance naming that session; attend all seven and the certificate reads Working with High Performance PostgreSQL, covering the whole track.
Are sessions recorded?
No. Sessions are live and interactive, with real-time Q&A. They are not recorded.
Can my company book this for a team?
Yes. Any session, or the whole track, can be delivered privately for a single team — online or onsite. Contact us at contact@dataegret.com.
Instructors
Over 15 years of experience across Oracle, DB2, MySQL, SQL Server and PostgreSQL. A recognised PostgreSQL contributor, active in the community through committees, user groups and international conferences.
Over 20 years of industry experience, with extensive expertise in supporting clients on their PostgreSQL journey and achieving high availability for business-critical projects.
A specialist in monitoring, high availability, performance tuning and scalable system architecture, with a background in banking, finance and telecommunications.
Ready to book?
Live, small-group, taught by the people who run PostgreSQL in production. Questions first? Write to us — no sales pitch.