From: Shyno Mathew (sm3334_at_columbia.edu)
Date: Wed Mar 05 2014 - 12:43:48 CST
We had to shutdown our cluster recently and this terminated all the FEP
runs. I haven't restarted an FEP before.
For each window, I have 10,000 equilibration steps followed by 20,000
production steps.My system was crashed for the lambda window 0.58 to
0.5825. From the .fepout file, it looks like only 850/10000 equilibration
steps were done.
Looking at the previous post, this is what I think I should do:
1. Run FEP for the window where it was crashed, here 0.58 to 0.5825
My restart frequency is 1000 fs. Since it crashed at 850, I assume I can do
the whole equilibration.
But I have other systems which crashed at 1050 fs, here I only do part of
2. Then use the restart files to run the remaining FEP, 0.5825 to 1 and
manually combine the .fepout files.
-- Shyno Mathew PhD student Department of Chemical Engineering Columbia University
This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:20:34 CST