Right now, you can’t do that on your end. We used to have a fixed number but we had to remove it because it was causing unintended issues. For instance, in some cases, someone would schedule 100s of runs for an account with limited resources. Then the runs would time out before even being scheduled. The best solution, for now, is to quality control your model and cancel them manually.
In the particular case of your models that are running g the UTCI comfort, the bottleneck is the energy simulation. It fails and then Pollination retries it again hoping that it can get it right next time.
I don’t know what would be the best approach in this particular case, as the models are very large and have too many shades for what EnergyPlus can handle. We might want to start thinking about adding GPU access to these steps to better handle the shade calculation.
You should probably try smaller models for a free account! Is it time to upgrade to a paid subscription?
Glad to hear that! Let us know if you have any additional feedback.