Continuing the discussion from Understanding Fly.io API Rate/Hard Limits:
My use case requires creating machines that’s above the documented rate limit (~10reqs/s to start with), but will grow in future. Just wondering if those are soft or hard limits?
Those are hard limits.
So to clarify your use case, you will be creating 10 Machines per second? Will you be destroying them at a similar rate? How many Machines do you expect to have running at any given time?
Thanks for the quick response Kyle!
The architecture I’m thinking is quite similar to Event Driven Machines where we spin up Fly Machines to process jobs coming off a queue.
Correct.
It depends how long the processing of the job takes, could be seconds or minutes.
Again it depends, but i’d expect somewhere between 5 and 10 to start with.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.