3.320  Atomistic Computer Modelling of Materials

LAB4 deadline is postponed from this Wednesday to next Monday (May 4). Once again, please limit the number of jobs per user per node to two, in order to allow everyone to complete the problem set. 

I realize that LAB4 is computationally intensive (many equilibration and production runs). Unfortunately the cluster is small and it is not possible to run too many jobs simultaneously. In particular, it seems there are a few users that are running several jobs at the same time in most of the nodes. In order to allow everybody to run their jobs and finish the LAB in time, please follow these 'suggestions':
- Never run more than two simulations per node. You can run the equilibrations and production runs one after the other with simple scripts (e.g. for i in Equil*; do cd $i; DL_POLY.X; cd ..; done). This rule is retroactive, if you are now running more than two jobs per node, kill them and resubmit them in serial. If you don't know how to do that, please ask me or tell me your username and I will do it for you.
- Carefully think at what you really need to simulate. Equilibration times can be much shorter ar low temperature, the high 'default' value in the script is more suitable when you are close to the phase transition. In any case, if you are running simulations now and you don't have much time to finish them I would not run more than 30ps of equlibration or production. Moreover, limit the number of different tempratures to the most significan ones: e.g in exercise II melting will probably occur around 300-350 K, run a few simulations in this range, some more at lower T (100, 200 K) and some more at higher T (400 K).

This is just to let you know that I added a FAQ.txt file in the /home/users/mit3320/LAB4 directory. It contains some of the most frequent and pertinent questions about this lab received so far. I will keep it up to date as new questions or problems emerge.

Some of the input files of Exercise Ib (160_tip4p.CONFIG, 240_tip4p.CONFIG, and 480_tip4p.CONFIG) have been modified/removed to avoid en error in DL_POLY ('quaternion integrator failed'). This is a subtle error, related to small deviations (close to the numerical precision) in the initial geometries of the water molecules. If you have not started yet this exercise, please remove the old /LAB4/ExerciseIb/Input and copy again the containt of  /home/users/mit3320/LAB4/ExerciseIb/Input in your /LAB4/ExerciseIb/ directory. Sorry for any inconvenience,
Oliviero

P.S. to be safe, before starting any exercise, copy the last version from the mit3320/LAB4 directory.

LAB4 handouts and problem set are now on stellar. They are slightly modified with respect to the printed versions, I added some references and put more details in the description of the scripts (and modified the due date in the problem set, it is not may 2007!).
Second point: if you have not started the lab already, please remove the LAB4 directory you created yesterday and copy it again from /home/users/mit3320/LAB4 . This is beacuse there were some mistakes in the directories (extracredits I and II were switched, extracredit III had a wrong input file) and in the run_ex* scripts. I would like to stress again that you are not required to use these scripts to run your calculations, but the are supposed to speed up the set up of the different simulations. Have a look at them, they should be selfexplaining. If you have any problem, please contact me (oliviero@mit.edu) or just show up in my office (13-4084). I apologize for any inconvenience.
Oliviero