3 ms·
None of these tests seem to run for longer than a minute? Drawing conclusions from experiments like this is fraught.
by natebrennand 1y ago
None of these tests seem to run for longer than a minute?
Drawing conclusions from experiments like this is fraught.
- abelanger 1y agoOP here, I agree. The point of the article is both educational and also aims to set a lower bound on latency and an upper bound on throughput for Postgres writes, and see how these tradeoff under different circumstances. I tried to make the testing framework easily accessible and tunable: https://github.com/abelanger5/postgres-fast-inserts https://github.com/abelanger5/postgres-fast-inserts For example, to run the benchmark for an hour: pg-inserts continuous copyfrom --duration 1h --batch-size 100 --max-conns 20 --writers 20 --flush-interval 10ms We run load tests on the order of hours and sometimes days at Hatchet -- which tends to squeeze our AWS bill but it helps us see behavior that we wouldn't otherwise see.