Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

One more quick question. Is there a way to rate limit flow runs?

you can set a concurrency limit on your work pool or the work queues in your work pool, which applies a maximum number of flow runs that can happen at any given time on that pool/queue

oh I just read your earlier post, you know this :facepalm:

Gotcha, but no means of offsetting start times by x time from the previous flow run?

hmm well I guess it depends on what's responsible for starting your runs

Mostly trying to prevent a "thundering herd" issue I'm encountering

Where many flow runs queue up from an event, kick off, then slam an API

I've partially resolved with with retries and jitter

but it would be great to be a bit kinder to the API :slightly_smiling_face:

you can use <https://docs.prefect.io/latest/guides/global-concurrency-limits/#using-concurrency-and-rate_limit-outside-of-a-flow|global concurrency limits>

Perfect, I think I have everything I need then

<@U05TQMN8RUP>: a clarification question about slot decay as there's a little room in the documentation for interpretation. Does the decay begin after the slot is initially occupied or after the slot is freed?

Also, in the UI I see a field called `Active Slots`  and in the Rest API an additional field called `denied_slots` that are not described in the documentation. Could I ask you for a little clarification of what those are and how to use them?