Cloud instances are too slow / timeout problems on larger instances

Hi all,
I have a fairly big deadline this Sunday for which I was hoping to run a non-toy-sized simulation: maybe 25 instances generating a meg or two of data each.

My problem is that, seemingly at random, instances are taking a very long time to spin up and run jobs. And when I go to a larger box, I am getting failed runs with timeout errors that look like this:

Note that the problem is not in my build, because the same build works fine under different circumstances (different box or different app_param).

Unity people in particular: is there anything you can do to help me ensure these runs can succeed in the next couple of days? Or tell me if not and I'll invest in getting docker up and running on one of my own machines.

Thanks!

 

We addressed this through simulation-help@unity3d.com. We identified an underlying issue in the service for long running simulations that has been fixed and deployed.

Thank you for your report.