Running Egs on WARP

egsnrc assumes you are working in tc-shell. Go into tc-schell say tcsh (to leave, exit)

egsnrc also assumes that you are working in a directory whose name is the same as the code you are working on. For example, if you're working on betaspec5.mortran, egsnrc expects you to be in a directory called ~/egsnrc/betaspec5.

to compile and link, from within that directory say

mf betaspec5

(no ".mortran" -- it's assumed, in fact it's required)

To execute code interactively say

ex betaspec5 "" pbcualscintsi1r

where the last parameter is a concatenated pegs4 datafile (pbcualscintsi1r.pegs4data) with all the materials, energy ranges, and other parameters that will be needed by egsnrc

Because I'm lame and doing UNIX scripts, they only work in bash so you must change to bash shell

To submit a batch job to run on a single processor say

qsub betaspec5bat.sh

in this case the number of trajectories is determined by the betaspec5.input file

To submit 10 batch jobs to run on 10 processors say

./parsub10.job betaspec5 10000000

where the number is the total number of histories. there's a parsub20.job also, but rarely that many processors free on WARP

Status of batch jobs may be checked using qstat

(to see how many processors are being used say "qstat -f | grep ppn" and count)

Progress (i.e. how many histories have been processed) can be checked by looking in the files in the root directory called baetspec5_0.tmp and so on. The first line contains the number of histories processed so far. It updates every 10,000 or so and when it's nearly done it will be 1/10 (or 1/20 if using parsub20.job) of the total number of histories specified.

After all jobs are finished, still in bash shell,

./combpar10.sh betaspec5

will combine the output results of the individual jobs into one bigger job, with generic names for the individual output files (e.g. newdat.dat etc.) as usual there is a combpar20.sh

To assign the aggregated generic output files to a run number say

cpnew xxxx

(where xxxx is the nun number).Now all the aggregated files that we keep you have that run number in them.

The output files created are

xxxx.dat which is an energy histogram of all events in the detector

xxxx.phdata which is an energy histogram of all events in the detector that are not associated with the positron also being in the detector

xxxx.*log contains a concatenation of the log files

xxxx.stat contains all the standard output information from running the code (including echoes of the the most important input parameters)

xxxx.22

xxxx.23 these are both diagnostics or trajectory tracing auxiliary files which are often turned off for speed.

  This Web Page Created with PageBreeze Free Website Builder