Synth is a tool for generating realistic data using a declarative data model

github.com

30 pointsfoolswisdom2 comments

I am grateful to the original poster for linking this here.

At my workplace we frequently benchmark and performance-regression test again PG + other SQL DB's and a tool like this has been sorely needed.

Many thanks to the authors for solving this real-world problem, GENERATE_SERIES() doesn't always cut it when you're looking to get an accurate estimation from a more realistic data model.

I'm glad you find it useful. I see from your profile you work at Hasura, I imagine this kind of testing is critical for them.

Personally, I was interested in Synth from the perspective of generating large mock data sets for use in a data pipeline.

Reply on news.ycombinator.com