Table of Contents
Setting up Environment and Invoking - Invoking CNSsolve on Emerald
Before accessing CNSsolve on Emerald, the following steps should be done. If setup for CNSsolve with ipm command (step 2) has been done before, skip steps 2 & 3. If you are not sure, use step 3 to confirm after login.
-
Open a terminal window, and type the following command at local machine prompt to log on to the Emerald cluster.
ssh emerald.unc.edu
-
At the Linux machine prompt, type the following command. Skip this step if you have done it before.
ipm add cnssolve
The shell scripts (public/.pkg.cshrc and public/.pkg.kshrc) in the home directory are then modified. The PATH to access CNSsolve and all necessary environment variables are set up properly.
-
To check if your startup scripts have been setup for CNSsolve, type the following command at the prompt.
ipm query
A line "cnssolve" is shown which indicates that the environment for CNSsolve is already set.
-
After the PATH and environment variables for CNSsolve are set up properly, CNSsolve can then be accessed. At the Linux machine prompt, type the following LSF command to invoke the software.
bsub -q week -R blade cns cnssolve.input.file cnssolve.output.file
where "blade" is one of the resource names on Emerald, "cnssolve.input.file" is the name of the input file and "cnssolve.output.file" is the name of the desired output file.
-
To use the MPI parallel form of CNSsolve, type
bsub -q week -n 4 -a mpich cns_mp cnssolve.input.file cnssolve.output.file
where 4 CPUs are used from the BladeCenter cluster.
Additional Help
[ http://its.unc.edu/research-computing.html ] Research Computing home page
Document Overview
Tags: CNSSolve, research computing servicesSubjects: Bioscience Computing
Genre: Tutorial
Audience: Everyone
This work is licensed under a Creative Commons Attribution 3.0 United States License.
Copyright 2002-2007 The University of North Carolina at Chapel Hill.
This
work is licensed under a
Creative Commons Attribution 3.0 United States License.
