PgBouncer: From 45K to 210K Queries per Second
A benchmark-driven look at how one PgBouncer process became a CPU bottleneck, how multiple processes scaled it, and where the next limit appeared.
- postgresql
- pgbouncer
- performance
- databases
Technical notes
Notes from building and benchmarking backend and data systems.
4 published notesAll notes
TurboVec on 100M RAG Vectors
A benchmark-driven look at what happens when vector compression moves from 10M scale to 100M scale.
- rag
- vector-search
- performance
1M Rows Into Postgres in 1.29 Seconds
A benchmark-driven look at how much the PostgreSQL write path matters when loading large datasets.
- postgresql
- performance
- data-engineering
Large JSON in Python: Parser Speed Is Not the Whole Problem
A benchmark-driven look at why large JSON ingestion in Python is often more about memory shape than raw parser speed.
- python
- json
- performance
- data-engineering