T
Tomas Vondra
vondra.me · Personal blog · on HN since 2024
Tomas Vondra, a PostgreSQL developer, writes detailed benchmarks and analysis of Postgres performance, I/O, memory tuning, and the Postgres community.
What HN saysHN links his benchmark posts, such as async IO tuning in Postgres 18 and BOLT optimization, which one commenter says gave gains of up to 40%.
20HN people shared it
2linked it in comments
21stories submitted
368story points
7 months agolast shared
2024people per year2026
Most-shared posts
- 1Tuning async IO in PostgreSQL 183 people · 101 points · first shared 2025 · 23 comments on HN
- 2Playing with BOLT and Postgres2 people · 167 points · first shared 2024 · 46 comments on HN
- 3Don't give Postgres too much memory2 people · 8 points · first shared 2025 · discussion on HN
- 4Fun and Weirdness with SSDs and PostgreSQL2 people · 6 points · first shared 2025 · discussion on HN
- 5The State of the Postgres Community2 people · 5 points · first shared 2024 · discussion on HN
- 6Benchmarking Is Hard, Sometimes2 people · 4 points · first shared 2025 · discussion on HN
- 7Postgres Performance Archaeology2 people · 3 points · first shared 2024 · discussion on HN
- 8How often is the query plan optimal?1 people · 32 points · first shared 2025 · 26 comments on HN
- 9Using JWT to establish a trusted context for Row Level Security1 people · 30 points · first shared 2025 · 14 comments on HN
- 10Will Postgres development rely on mailing lists forever?1 people · 5 points · first shared 2024 · discussion on HN
- 11So why don't we pick the optimal query plan?1 people · 3 points · first shared 2025 · discussion on HN
- 12Postgres Performance Archaeology: OLAP1 people · 2 points · first shared 2024 · discussion on HN
- 13Stabilizing Benchmarks1 people · 1 points · first shared 2026 · discussion on HN
- 14The real cost of random I/O1 people · 1 points · first shared 2026 · discussion on HN
What HN people wrote when they linked it
mebcitto · 2025-10-24
That explains why `sync` and `worker` have so similar results in almost all runs. The benchmarks from Tomas Vondra (https://vondra.me/posts/tuning-aio-in-postgresql-18/) showed some significant differences.
JoelJacobson · 2024-11-06
Here is another interesting BOLT article, this one on PostgreSQL optimization:
https://vondra.me/posts/playing-with-bolt-and-postgres/
"results are unexpectedly good, in some cases up to 40%"