Half a dozen exciting new features in PostgreSQL 16, 17 and 18
Presented by:
Thomas Munro
I'm based in Wellington, New Zealand where I've been working full time on PostgreSQL for a decade, first for EnterpriseDB and currently for Microsoft. I am an active developer and committer on the project, and frequently speak at the major developer conferences overseas about projects I contribute to. My areas of work have included parallel query execution, I/O, recovery/replication, portability and standards, modernisation and ongoing maintenance.
Before achieving open source escape velocity, I did database, network, application, embedded and systems programming for a couple of decades, mostly in London but also for a few years in Paris, Sydney and Wellington for employers in the finance/trading, electricity, education and advertising sectors.
The other tech-related project I'm involved in is the FreeBSD operating system, where I have made small contributions primarily relating to PostgreSQL performance, and continue to experiment with the database/OS interface layer. For non-tech interests, I'm a failed linguist and student of Western European languages, occasionally a road cyclist, and am currently trying to learn to surf... by Australian standards very poorly, despite which it's always nice to visit NSW and be reminded of that.
This is a three part talk. The first part gives a very high level overview of the areas of development in recent major releases, including planning, execution, partitioning, replication, backups, monitoring, security, I/O, ongoing modernisation work and more.
In the second part, we'll look more closely at three selected new features that bring everyday benefits in terms of usability and maintenance:
- extended standardised JSON support
- new text collation support, and why it matters to you
- fast and efficient incremental backups
In the third part, we'll talk about three selected major internal changes that improve performance and bring us closer to the state-of-the-art:
- overhaul of the statistics and monitoring system
- btrees with skip scans and related tricks to get more benefit from your existing indexes
- modernised high performance disk I/O and streaming, an ongoing project
This covers work done by a lot of different people and companies in PostgreSQL 16, 17, 18 and beyond. I've personally been involved in developing the I/O layer along with my team at Microsoft and many others in the community, among other projects.
- Date:
- 2025 October 17 09:00 +11
- Duration:
- 40 min
- Room:
- Oxford 1 + 2
- Conference:
- PG Down Under 2025
- Language:
- Track:
- Essentials
- Difficulty:
- Medium