...
If you are uncertain of a suitable approach, please contact cluster-support@tufts.edu for assistance.
Ansys and Fluent related:
...
PetSc related.
How do I run Ansys WorkBench interface?
-bash-3.2$ module load ansys
-bash-3.2$ srun ... runwb2
How can I read Ansys/Fluent/WorkBench and other docs while on the cluster?
You may run the Ansys Help application after you set up the module by:
| -bash-3.2$ module load ansys/16.2 -bash-3.2$ srun --pty --x11=first -p interactive -c 1 anshelp162 |
|---|
Documents are version specific.
How do I run ansys in batch mode without using the Launcher if I already have an Ansys .dat file prepared?
> module load ansys/15 > sbatch --wrap="ansys150 -c 4 -j vm233 -b < vm233test.dat" |
|---|
Where can I find example ansys files? Various directories for different versions are available.
/opt/shared/ansys_inc/v130/ansys/data/models/
The following Fluent related info may no longer work due to Ansys incorporation of Fluent into WorkBench.
How do I run a Fluent 2D parallel job?
Define the number of cpus(up to 8) within the bsub and fluent command line. The -n and -t arguments is the location to change the number of cpus.
-bash-3.2$ sbatch ....sbatch_options --wrap="fluent 2d -g -t8 -pib -mpi=intel -i yourfluent.script"
Note: replace 2d with the appropriate fluent product, such as 3d and so on...
- What is a useful Fluent resource?*
CDF-Online
How can I view the Fluent docs?
> module load fluent
> firefox file:///opt/Fluent.Inc/fluent6.3.26/help/index.htm
How do I get fluent command line options?
-bash-3.2$ srun ... fluent -help
How do I run the 2D or 3D fluent without asking the questions interactively?
-bash-3.2$ srun ... fluent 3d
How do I run Icepak?
Icepak is available from within Ansys WorkBench interface and as a command line standalone application. Start WorkBench in the normal fashion, as above. Icepak will be a choice on the Components Pane, click on it and then click setup. To run just the Icepak interface try:
-bash-3.2$ module load ansys
-bash-3.2$ srun ... icepak
Abaqus related:
How do I run Abaqus?
There are several ways to run jobs. Most will find the following useful for checking syntax,etc..:
-bash-3.2$ module load abaqus -bash-3.2$ srun ... abq682.exe job=mydesign datacheck interactive input=./design.inp |
|---|
To actually run the code, remove the option datacheck and interactive.
This will run your abaqus input file called design.inp as a serial job and if successful will create about 14 files in the directory you are working in. One such file is the .odb Abaqus database file. All the files are created with the job name "mydesign" and differ in the filename extensions and type of file.
Note, normally there are several versions of Abaqus available. Therefore one needs to note which one reflects what was set up with Modules. For example, the command abq682.exe is version specific.
How do I run a parallel Abaqus job?
| -bash-3.2$ srun ... abq682.exe job=design_4cpus cpus=4 mp_mode=mpi input=./design.inp |
|---|
Note different Abaqus products may invoke different solvers. Check the documentation for options.
How can I find the abaqus command line options?
Suppose you are using the default version of abaqus.
> module load abaqus >srun --pty --x11=first -p interactive abq6112.exe -help This will produce a several page list of command line options. |
|---|
How can find out what abaqus licenses are available?
> module load abaqus > srun --pty --x11=first -p interactive abq6112.exe licensing lmstat -a -c |
|---|
How do I run Abaqus version 6.9.2 CAE gui?
| -bash-3.2$ srun ... abq692.exe cae |
|---|
Why is OpenGl not working with CAE?
Add the -mesa option to the start of abaqus as mentioned above.
How do I read Abaqus documentation?
If you have X11 forwarding in place you will get a webpage with documentation.
-bash-3.2$ module load abaqus -bash-3.2$ srun ... abq682.exe doc |
|---|
Comsol related:
How do I run Comsol as an interactive threaded job?
| > module load comsol/5.0 > salloc -N 1 -c 8 -p interactive > srun --pty --x11=first comsol |
|---|
How do I run Comsol as a mpi parallel job?
Start Comsol as above, but add the command line options as follows:
| -bash-3.2$ srun ... comsol -np 8 -mpmode owner |
|---|
Then in the gui setup for Solver parameters, change the default Solver to the Parallel Direct Solver(Pardiso) and rerun your model.
To do the same task but on an node with exclusive access use the --exclusive option, but note that the request may take some time to be granted depending of loads.
The comsol help button doesn't work, how do I find the documentation?
The compute nodes do not have browsers installed. Firefox is available on the headnode of the cluster instead. The install tree for Comsol contains html docs and some pdfs as well.
$ ls /opt/comsol35a/doc/
aco browser chem multiphysics plain sme
For example to view the html chem docs:
$ firefox file:///opt/comsol35a/doc/chem/index.htm
Sometimes you may find that your comsol efforts no longer work as they may have in the past; resulting in java errors of several types. These errors will be noted in your log file. What can one do to get around this?
Close any open Comsol sessions, and remove your .comsol directory located in your home directory. Make a backup and then delete everything in .comsol directory. > mv .comsol .comsol.backup > mkdir .comsol > ln -s /cluster/shared/your_utln/.comsol ./.comsol |
|---|
This soft link will redirect comsol temp structures to your dir. in /cluster/shared/ . There is no disk quota on this filesystem.
Restart comsol and a new set of structures will be created. If you have requested Research Storage, you may use that for a longer term solution.
Note: You only need to do this once.
How do I run comsol in batch mode?
> module load comsol To run Comsol's tube1b.mph example: > sbatch ...options.... --wrap="comsol batch -inputfile tube1b.mph -outputfile tube1b-output.mph" |
|---|
The solution will update a new version of the .mph file as output. You can read the output file into Comsol later for further post processing. There will be 3 output files and a directory created with name tube1b_files containing graphics and other files produced from the batch job.
Note, you may then log off while your job runs. YOu will be notified by email when the job is done.
How do I run comsol gui interface with software rendering?
Use the Comsol commandline option: -3drend sw
| > srun --pty --x11=first -p interactive comsol -3drend sw |
|---|
This may solve display issues you might have do to your xserver.
GAMESS related:
-bash-3.2$ module load gamess
-bash-3.2$ srun ...options... rungms exam02
Imagemagick related:
Where do I find information on Imagemagick graphic tools?
Imagemagick home, usage info and scripting links. Also local man pages on individual commands is available. Note Imagemagick is not a command. It is a suite of programs.
For example to use the convert program:
-bash-3.2$ module load ImageMagick How to display a picture using Imagemagick's display command without running the interface |
|---|
Matlab related:
- How do I get matlab command line options?*
-bash-3.2$ srun ... matlab -help
How do I run a matlab batch job?
-bash-3.2$ module load matlab -bash-3.2$ sbatch ...options... --warp="matlab -nojvm -nodisplay < your-matlab-code.m" |
|---|
This will submit your code to the slurm partition you choose in the sbatch options. Matlab has several ways to direct output and results by use of matlab commands such as: diary, print and several other commands.
How do I speed up my Matlab calculations?
By default Matlab on the cluster has multi-thread parallelism enabled. Many Matlab functions will automatically use the threaded libraries. The number of threads obtained is equal to the number of cores on the node your job is placed on. Currently it is either 8 or 12. You no longer need to request exclusive node access via bsub.
How do I access the Parallel capability of Matlab?
Matlab can make use of MPI parallelism if you modify your code accordingly(see Parallel computing documentation within Matlab) and setup your session to instruct Matlab to use slurm. Currently Distributed Computing Engine Toolbox is licensed for 32 cores. It is strongly recommended to initially use 4, 8 or 16 cores while you explore how your code scales. Depending on what you are doing, execution times may not benefit from additional cores. To setup this feature, import a working configuration found at:
----Currently this has not been tested with slurm. Matlab R2015b, expected Fall 2015 will have slurm support -----
To view an older lsf based config file:
/cluster/shared/dmarshal/ISO/matlab-R2012-lsfconfig16.settings
Use Matlab's Cluster Profile Manager Tool to import the settings. Choose the menu options Parallel -> Import Cluster Profile and browse to the config file.
To test this out, you may run the Validation tests by choosing Start Validation. This will take several minutes and should result in five passed tests.
I want to run hundreds of matlab jobs using my matlab scripts. What do I need to know?
Your jobs will end up on a partition with a limit of 512 concurrent jobs/cores/user.
Is Matlab MCR available to run compiled jobs?
Yes. You will need to load the MCR version corresponding to the Matlab version you used to compile your standalone executable. The Matlab example called magicsquare in the documentation can be used as an example. Once compiled the named executable, magicsquare, can be run without loading Matlab nor checking out licenses. Such an example is available below. Note, be sure to quote any needed command line arguments to your named executable.
> module load MCR/R2012a > srun ... "/cluster/shared/dmarshal/ISO/magicsquare 3" |
|---|
How do I access the public domain Matlab toolbox EDGE?
EDGE has been installed into cluster directory /opt/shared/EDGE/1.02. To use it you'll need to launch MATLAB then change into the /opt/shared/EDGE/1.0/Matlab directory by issuing the following command at the Matlab prompt:
cd /opt/shared/EDGE/1.0/Matlab
Or better yet, add this path to your Matlab path within Matlab. Choose File -> Set Path and Add with subfolders.
To launch the Importer tool from within Matlab type on the command line:
semiauto
See the EDGE website for additional documentation.
How do I access the public domain Matlab toolbox WaveLab?
Load the corresponding module:
> module load Wavelab
> module load matlab
In Matlab choose File -> Set Path and then Add with Subfolders. Add the following: /opt/shared/Wavelab/Wavelab850
Then issue the command startup at the Matlab command line.
At this point you can issue commands and demos. Check the website for documentation.
Mathematica related:
Where can I find more Mathematica documentation?
The vendor's website and the online help in the application. In addition look at the broadcast section on the vendor site to view various movies on how to do many tasks in Mathematica. Also check out some paper based options.
My Mathematica notebook file doesn't work like it did before, what is up?
The following explanation is a good starting point to explore this issue.
How does one submit a mathematica batch job?
Suppose you have a text file called my_mathematica.m created with a text editor with the following tasks:
AppendTo[$Echo, "stdout"] To submit as a batch job through the batch partition, but not the Mathematica gui interface: |
|---|
I get an error about Mathematica missing fonts,etc... What does this mean?
The cluster does not provide a Font Server supporting Mathematica font sets. Whereas a local install of Mathematica on your desktop includes this font set. Remote users need to provide access to the fonts locally.
For linux desktops, the easy way is to obtain the Mathematica helper program called, Mathematica Player. You may download this from the vendor website. Before you start Mathematica on the cluster for display on your computer, start the helper program locally. This will provide the font support.
For MacOSx desktops you need to either install Mathematica or the Mathematica Player and then point to the installed fonts. For example if you have Mathematica installed, to use it's fonts, open a terminal window(shell) and type:
> xset fp+ /Applications/Mathematica.app/SystemFiles/Fonts/Type1/
> xset fp rehash
Assuming that is the path to the fonts.
Then connect to the cluster and start Mathematica normally.
NCAR related:
How do I start NCAR?
NCAR has lots of separate programs. These are all available to you after you setup the environment.
-bash-3.2$ module load NCAR
These executables are located at: /opt/shared/NCAR/5.2.1/bin
To submit a job, for example:
-bash-3.2$ sbatch ...options..... --wrap="ng4ex gsun01n.ncl"
Please consult the NCAR website for user documentation and examples.
Python related:
How can I verify if a particular Python package is installed?
Add-on tools such as numpy and scipy are installed. Others would be under the install tree located at:
/opt/shared/python/
in the version specific site-packages directory. Another approach uses pip.
| > module load python/2.7.6 > pip list |
|---|
PetSc related.
How do I install PetSc?
PetSc has a complex build environment that allows one to include various applications. We encourage interested users to build the suite of interest in their home directory. Installation instructions for a particular set of applications can be found here.
Perl related:
How can I verify if a particular Perl package is installed?
Suppose you wanted to see if package GD is installed:
-bash-3.2$ rpm -q perl-GD gd
PDF related:
How do I view pdf files on the cluster?
Use the X11 based program evince. It is already in your PATH.
R related:
How do I run a long running R job as a batch job and I require lots of ram?
You may send your R script to the slurm batch partition requesting access to a 32gig ram node, as an example:
| -bash-3.2$ sbatch --mem=32g -p batch --wrap="R CMD BATCH your-inputfile.r outputfilename" |
|---|
For example, for the most part, option -p batch will place the job on nodes with 128gig memory allocations. Default units of memory is megabytes:
--mem= 30000 will request 30gig of ram. Use of the largemem partition should be considered when more total ram is required.
Where is the R documentation?
Extensive user documentation and tutorials are available on the R web site. There are many texts as well, here is nice example
And if you are an Spss or SAS user check this out.
And another source for comparative statistical package example usage can be found at UCLA
Where can I find command line R documentation?
> module load R
> srun ... R CMD INSTALL --help
Or use the man pages available at the linux shell prompt.
-bash-3.2$ man R
Also of note is the following online CodeSchool tutorial
Additional info on R can be found in the admin guide
How can I install an R package that is only available as a tar file?
Download the tar file and from within R try the following. Make changes to the utln and path to reflect the location. For example:
| > install.packages("/cluster/shared/utln/your_R_package.tar", repos = NULL) You may also install a package into another location: -bash-4.1$ srun ... R CMD INSTALL --library=/cluster/shared/utln /cluster/shared/utln/someRpackage.tar.gz Change the utln to your login name and adjust the filesystem path. |
|---|
You will need to add the location to R search path for libraries.
Within R type:
> .libPaths() This is the default library system wide. Now to add the new additional location: > .libPaths(/cluster/shared/utln/) > .libPaths() |
|---|
How can I install an R package from the CRAN repository?
From within R issue the commands:
> options(CRAN="http://cran.us.r-project.org/") Note that R package names are case sensitive. Alternatively, try: |
|---|
Edit the utln to reflect your Tufts username and directory locations. Locations may be home directory or optional storage locations.
Check the R documentation for additional options. Note, you home directory quota is limited to 5 gig.
Stata related:
How to choose between SE, IC and MP editions?
Most users will likely only need IC. All versions have the same statistical functionality but differ in the size of their data structures.
How does one send a Stata batch job to the cluster compute nodes?
Make a stata do file with stata commands. To send:
| > sbatch --wrap="stata -b your_filename.do" |
|---|
How to run stata interactively:
bash-4.1$ srun --pty --x11=first bash bash-4.1$ module load stata/13 bash-4.1$ xstata |
|---|
What are the command line arguments to Stata?
To find out:
| > srun --pty --x11=first -p interactive stata -help |
|---|
SAS related:
To run interactively:
> module load SAS > srun --pty --x11=first sas_u8 |
|---|
For batch use:
> module load SAS >sbatch -p batch sas_u8 yourprogram.sas |
|---|
You will get yourprogram.log and yourprogram.lst output files dumped to your current directoryinstall PetSc?
PetSc has a complex build environment that allows one to include various applications. We encourage interested users to build the suite of interest in their home directory. Installation instructions for a particular set of applications can be found here.
Perl related:
How can I verify if a particular Perl package is installed?
Suppose you wanted to see if package GD is installed:
-bash-3.2$ rpm -q perl-GD gd
PDF related:
How do I view pdf files on the cluster?
Use the X11 based program evince. It is already in your PATH.
Tecplot related:
On the cluster Tecplot's help button doesn't work, where do I find TecPlot documentation?
Html docs can be found on the vendor's website
PDF versions are an option as well.
...
What is required to run Solar?
This program requires a user specific ID key. Send a request to cluster-support@tufts.edu requesting one. This can take up to 24 hours for the vendor to respond. This is a one time registration event. Once you have received your key:
...
This example is straight forward. However enhancements to bsub options and to QIIME python functions will vary as will data sources locations.