| Linux Cluster Installation and VLS. | 
ICM can be configured to run on a linux cluster.
You need the following components:
- A linux cluster each node of which mounts the ICM home directory, 
your chemical database directory, and the project data.
- Floating FLEXLM licenses to start the needed number of simultaneous ICM processes, 
or, alternatively, nodelocked licenses for each node in the cluster you are going to use.
- The  PBS  queuing system to submit VLS jobs to the cluster. 
If you use another queuing system, convert the PBS scrippts to a different queuing program.
To run a PBS docking job do the following:
- customize your rundock shell script (it should work as is if the connectivity and mounting is OK)
- customize and test your pbsrun  script that queues a single rundock process to PBS and passes
parameters to rundock . Example:  qsub $ICMHOME/pbsrun -vJOBARGS="-f 1 -t 10000 DOCK1"  
- customize and test your pbsscan  script that queues multiple rundock processes to PBS by dividing
the compounds into chunks. Example: 
pbsscan DOCK1 1 1000000 10000  # queues 100 rundock processes each working on 10000 compounds