Is the DCD file size consistent for consecutive runs?

From: Bjoern Olausson (namdlist_at_googlemail.com)
Date: Wed Oct 06 2010 - 03:36:04 CDT

Hi NAMD users,

just a simple question:

Can I expect the _exact_ same DCD file size for consecutive runs whit all the
same parameters (except of the increasing firsttimestep)?

run1: 50000 Steps
run2: 50000 Steps (continued from run1)
run3: 50000 Steps (continued from run2)
---> 150000 Steps (for all runs)

Is the following statement true (in terms of file size)?
run1.dcd + run2.dcd + run3.dcd == 3*run1.dcd

You wonder why I am asking this. The answer is simple:
I was running a simulation splitted in 1ns pieces with a python script which
handled the restarting and did some sanity checks like comparing the number of
steps written to the restart file to the expected steps. Nothing special
occurred during the runs (343 so far).
Yesterday I started some analysis and discovered that four of the trajectories
were broken:

Size Run
1524552 run051 <- O.K.
 396388 run052 <- broken
 396388 run142 <- broken
  19828 run232 <- broken
  51844 run322 <- broken
1524552 run323 <- O.K.
1524552 run324 <- O.K.

But for all runs, the log showed the correct number of steps written and the
restart.xsc file contained the expected steps. Only the DCD file is short of
steps.

I doubt that moving the file to another directory for storing (on the same
partition) is the cause since shutil.move() is atomic for files on the same
partition.

So I decided to add another sanity check: File-size - since there were no
other hints for failures (Normal program termination, normal count of steps in
the output, and normal count of steps in the restart file).

checksize(new, old):
  if os.path.getsize(new) == os.path.getsize(old):
    return True
  else:
    return False
        

Did anyone experience something similar?

Cheers
Bjoern

-- 
Bjoern Olausson
Martin-Luther-Universität Halle-Wittenberg 
Fachbereich Biochemie/Biotechnologie
Kurt-Mothes-Str. 3
06120 Halle/Saale
Phone: +49-345-55-24942

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:34 CST