Show HN: Django-xbench – slow endpoint aggregation for Django(github.com)

11 pointsby yeongbin055 miesięcy temu5 comments

Show HN: Django-xbench – slow endpoint aggregation for Django

https://github.com/yeongbin05/django-xbench

5 comments

blorenz5 miesięcy temu

Terrific project. I use Strawberry GraphQL so about all of my endpoints will be /graphql unfortunately. It would be nice if it could parse off a header to qualify which bucket it goes to. That way I could send over the operation and django-xbench takes care of it.

I see you are going have N+1s on the roadmap. I currently use django-zeal which does a fantastic job detecting them but there isn't a visualization. I'm keenly watching where you go with this.

I need more django perf in my life.

yeongbin055 miesięcy temu

Thanks for the thoughtful feedback!

I haven't planned GraphQL-specific grouping yet. N+1 visualization is the current priority but I'll definitely keep this in mind. Thank you very much

patrick915 miesięcy temu

how are you finding Strawberry?

simonw5 miesięcy temu

Gathering the stats in-memory is neat - I'd expected this would be writing to a DB but that's not how it works: https://github.com/yeongbin05/django-xbench/blob/f63316126b5...

yeongbin055 miesięcy temu

Previously it focused on per-request timing. This update adds rolling-window aggregation to detect performance trends.