Software available on the CIBR Clusters
Most tools available on standard Linux installations will be available:
- GCC collection
/usr/local is shared among all of the cluster nodes. So most extra software will be installed within that folder. You can find a lot of available tools simply by checking /usr/local/bin.
Here is a list of some other tools we have currently installed for users on specific clusters:
Installed in /usr/local/EMAN2. Updated at irregular intervals. See below for details.
R is installed on all of the clusters with various add-ons requested by users. In general it should be possible for us to install any non-commercial R packages if multiple users require them or they cannot be readily installed in individual user accounts.
BCM has a site-license (organized by CIBR) for Matlab and several toolkits, BUT this license is for use on individual PCs. The official Matlab cluster toolkit is EXTREMELY expensive (something like 20-30% of the cost of the cluster itself), and we simply do not have the budget for this sort of purchase, given that only a very small minority of cluster users use Matlab.
It IS still possible to run Matlab on the clusters, but in theory you can only use it one node per job (16 or 24 cores), and you have to jump through a few hoops with licensing to get things up and running. As it says on the website above, Larry Mayran (firstname.lastname@example.org) is BCM's Matlab license administrator, and he should have a bit of information on how to do this. We could also put you in contact with a couple of other cluster users who have solved some of these problems, but for both legal and manpower reasons, the cluster managers can't officially help you to use Matlab on the cluster. If you attempt installation yourself, and run into some sticking point you have been unable to find a solution to, we may be able to offer some advice, but we don't have people who can take you through the process step by step.
The cluster-installed version of EMAN2.1 in /usr/local/EMAN2 is optimized for the cluster, giving it a 10-20% performance boost over the binaries you might download from the website. It has also already been configured to work with MPI on the cluster. Set the following environment variables in your .bashrc file to use it:
export EMAN2DIR=/usr/local/EMAN2 export LD_LIBRARY_PATH=$EMAN2DIR/lib:/usr/local/lib export PYTHONPATH=$EMAN2DIR/lib:$EMAN2DIR/bin export PATH=$EMAN2DIR/bin:$EMAN2DIR/examples:/usr/local/bin:$PATH
You will also need to make sure you have a .bash_profile (which is used instead of .bashrc for batch jobs) containing something like:
# Get the aliases and functions if [ -f ~/.bashrc ]; then . ~/.bashrc fi
To run jobs using MPI on the cluster, run e2refine_easy.py or other program using the --parallel option. For example, if you requested 96 cores from the batch system, you would specify --parallel=mpi:96:/scratch/<username>. You should NOT issue the mpirun command yourself. This is done for you by the individual programs based on the --parallel option.
Additionally, for programs accepting it, you should specify the --threads=<N> option. Here N is NOT the number of MPI tasks, but is the number of threads available on each node. On sphere, 30 is a good number, assuming you have used the 'bynode' queue. On Prism, 24 is a good number to use. While there are only 16 physical cores present on Prism nodes, the machine can gain a modest (10-20%) performance boost by launching about 50% more threads than physical cores.
Remote GUI display from the cluster is not generally feasible, so we suggest limiting yourself to running large compute jobs on the cluster, and doing any GUI work on your local machine. The easiest way to handle this is to simply rsync the full project directory between machines. For example, consider this session with a local workstation and the sphere cluster:
local> pwd /home/stevel/myproject local> rsync -avr ../myproject sphere:data .. rsync output not shown local> ssh sphere sphere> cd data/myproject sphere> vi myslurmjob.sh edit job file sphere> sbatch myslurmjob.sh wait for job to complete exit sphere session local>pwd /home/stevel/myproject local>rsync -avr sphere:data/myproject .. .. rsync output not shown local> e2projectmanager.py examine refinement results