Lester Hedges

Not that I don't want to add tests in future, so please feel free to do this within the sandpit as required. Part of the issue is the size of...

Just to clarify, you mention binding free energy calculations, but are seeing poor performance solely for an equilibration? Thoughts / questions below: * Could you post a Gromacs log file...

Interesting, I tried explicitly adding `-nb gpu` on our cluster but it made no difference. (Perhaps it won't for a small system.) When I looked at [this](http://manual.gromacs.org/documentation/current/user-guide/mdrun-performance.html) page it seems...

For reference, here are the relevant sections from a GROMACS log for one of the free legs of an ethane-methanol perturbation on BlueCrystal 4: ``` GROMACS: gmx mdrun, version 2018...

I also noticed that our GROMACS version isn't compiled to use `thread_mpi` as its MPI library. According to [this](http://manual.gromacs.org/documentation/5.1/user-guide/mdrun-performance.html#running-mdrun-within-a-single-node), I shouldn't expect to get as good single node performance as...

No, my results are for the actual lambda = 0 stage of the free leg, so it's not a direct comparison. I was just showing how GROMACS auto hardware detection...

Changed to a more appropriate issue title. From poking around it seems like this isn't a specific BioSimSpace problem. We should use this thread to debug and document reliable ways...

Thanks for reporting, that's good to know. Since we don't bundle a version of GROMACS it's hard to provide settings that are optimised for any version (and hardware environment). I'm...

Thanks, @annamherz! I should be able to take a look at this in detail early next week.

Sorry for the delay with this. I'll try to make some comments today/tomorrow. Don't worry about making any edits before the workshop, we can deal with them afterwards when we...