Running Scale
- Log into any DECF Linux clients
- To run SCALE, use one of the following commands:
batch5 [-m -r -t -p -n N] inputfile [outputfile]
where [
-m = print messages from KENO, XSDRN, MORSE, NEWT to the screen (optional flag)
-r = don't remove temporary working directory at end of job (optional flag)
-t = use specified existing temporary working directory (optional flag)
-p = print block letter banner pages in output (optional flag)
-n N = nice value to be used during execution (default N = 2)
]
inputfile = input file name (required) (can input path)
outputfile = output file name (optional) (can include path - all output files
will go to the same directory as the output file) (if not specified,
the output file will be drived from the input file, and output
will be returned to the input file directory)
scale5 inputfile [outputfile]
where inputfile is the SCALE input file without the extension. This
script works with input file names with no extension (e.g., case1), or
extensions of .inp, or .input (e.g., case1.inp, case1.input). The
output file will automatically be written with the .out or .output
extension (e.g., case1.out or case1.output), respectively. The batch5
script also automatically saves system messages in a .msg or .msgs
file (e.g., case1.msg or case1.msgs), respectively. The batch5 script
automatically sets the environmental variables needed by the scale5
script, allocates a temporary directory for the calculation, calls the
scale5 script, and deletes the temporary directory after scale5
completes execution. If you use the scale5 script instead, you must
set the required environmental variables and allocate the temporary
directory. The directory will not be deleted when execution is
completed. The environmental variables that must be set are:
SCALE - the full directory path where SCALE is installed
TMPDIR - the full path for the temporary directory
After executing the problem using the scale5 script, output files are
collected and returned to the user's directory. If there is a problem
with execution, sometimes the files are not returned. The user can
manually go to the temporary directory and examine the files, print,
_out000n and _prt000n, etc., to find the problem. The files beginning
with _ are numbered with an integer that increments each time another
file is written.
- For Example
unix# batch5 -r input.inp
e.g. scale5 test.inp
This will save your working directory after the job is done.
Note that default output filename for batch44 is filename.output.
OR
unix# scale5 input.inp output.out
e.g. scale44 test.inp /tmp/test.out
You should put your output file in the /tmp/ directory if it will be large and you will go over your disk quota.
| |
