Atlas Parallel Running Quota

I’m working on graduate research related to the reliability of hyperscale cloud platforms. I use Ripe Atlas to monitor network reachability using many short duration ping and DNS measurements. With only 1-2 probes, these tests complete in < 30s typically.

I began hitting the parallel running measurements quota (max 100). From what I can tell, it seems that once you start a measurement, there is a minimum time it will count against the quota - maybe in the 5-10 minute range. Is there any guidance on what this duration is?

So the notional duration for a oneoff is indeed 5 minutes, but that’s complicated by the fact that the process which “stops” them only runs every 5 minutes, which gives you the 5-10 minute range. We have an upcoming change that will reduce that to 5-6 minutes.

If you’re willing to share details of your research campaign with me (camin at- symbol ripe.net) then we can discuss ways of making it work better, potentially temporarily increasing your concurrent measurement limit.