Matlab

From UFRC
Revision as of 16:36, 18 April 2017 by Jakers (talk | contribs)
Jump to navigation Jump to search

Description

matlab website  

MATLAB is a programming environment for algorithm development, data analysis, visualization, and numerical computation. Using MATLAB, you can solve technical computing problems faster than with traditional programming languages, such as C, C++, and Fortran.

You can use MATLAB in a wide range of applications, including signal and image processing, communications, control design, test and measurement, financial modeling and analysis, and computational biology. For a million engineers and scientists in industry and academia, MATLAB is the language of technical computing.

Restrictions

Because we are restricted to using only compiled versions of MATLAB programs, we are restricted to which toolboxes can be incorporated and used on the system. The full list of ineligible toolboxes is on the MATLAB website:

MATLAB Ineligible Programs

System Variables

  • HPC_{{#uppercase:matlab}}_DIR - installation directory



Policies

  1. You may not run MatLab nor compiled MatLab executables interactively on the login servers.
  2. You must use a development SLURM session to test and compile your MatLab programs.
  3. Long running (more than 30 minutes) MatLab jobs must be submitted to the batch system.
  4. UF Research Computing has no budget for application software and licensing. All MatLab licenses (including the compiler) come from, and are shared with, all of UF campus. For this reason it is sometimes difficult to obtain a license or MatLab may take a long time to start. If your research depends on MatLab and you do not want to be delayed by licensing issues, you should consider purchasing a MatLab license for your personal or group use.

Available components

See the Matlab add-ons for a full list of licensed Matlab components.

Compiling MatLab Programs

MatLab programs must be compiled via the MatLab compiler before they can be run in batch mode on HiPerGator.

In order to compile MatLab programs you should

  1. See Development and Testing page to learn how to get an interactive developmental session on HiPerGator under SLURM.
  2. Set up your environment so that the libraries needed by the compiler can be found.
module load matlab

Note: The module load command should be included in your batch submission script as well since your compiled matlab executable will also need to find the appropriate dynamically loaded libraries.

  1. Compile your matlab program (.m) files using the MatLab compiler (mcc).
mcc -R -singleCompThread -m some_prog.m

Example

Let us say that we have a program such as that below in a source file named houdini.m

function [m,t] = houdini();
% Doc example, chapter 3.

% Copyright 1997 The MathWorks, Inc.
% $Revision: 1.1.6.1 $

m = magic(2); % Assign 2x2 matrix to m. 
t = m .^ 3;   % Cube each element of m. 
disp(t);      % Display the value of t.

Since we already have the environment setup, we can compile this program. It is recommended that this be done in a separate directory as a number of files are produced.

 mkdir tmp
 cd tmp
 cp ../houdini.m .
 module load matlab
 mcc -R -singleCompThread -m houdini.m 
 ls -l
total 124
-rwxr-xr-x  1 root root 12299 Apr  3 10:08 houdini
-rw-r--r--  1 root root 63146 Apr  3 10:08 houdini.ctf
-rw-r--r--  1 root root   239 Apr  3 10:06 houdini.m
-rw-r--r--  1 root root  2900 Apr  3 10:08 houdini_main.c
-rw-r--r--  1 root root  6328 Apr  3 10:08 houdini_mcc_component_data.c
-rw-r--r--  1 root root  5461 Apr  3 10:08 houdini.prj
-rw-r--r--  1 root root  1010 Apr  3 10:08 mccExcludedFiles.log
-rw-r--r--  1 root root  6757 Apr  3 10:08 readme.txt
-rwxr--r--  1 root root  1446 Apr  3 10:08 run_houdini.sh

The MatLab executable houdini has been generated, and can be run by typing ./houdini, or by putting that command in a batch submission script.

Note: The "-R -singleCompThread" options on the mcc command line will result in a single-threaded executable. By default, multi-threading is enabled in executables produced by mcc. The result is that one computational thread will be started for each processor in a machine that runs your executable. That's not usually what you want. Unless you know what you are doing and intend to take advantage of multithreading (including requesting the appropriate resources in your submission script) you should use these options and generate a single-threaded executable.


MatLab Compiler Documentation can be found on the MatLab Website.

MATLAB GUI session

Start a MATLAB GUI session by loading the GUI module:

module load gui

Use the MATLAB launch wrapper to start your MATLAB GUI session:

launch_matlab_gui

If you would like to pass custom parameters to the job script, these parameters can be listed using the -h parameter:

[$USERNAME@gator3 ~]$ launch_matlab_gui -h

Usage:

        launch_xpra_gui_$application [options]
                                                                                                      
Options:                                                                                              

        -h - show this help message

        -e <executable> - program to run (REQUIRED)

        -m <memory> - memory, gb (default is 4gb)

        -p <procs> - processor cores, (default is a single core)

        -t <time> - SLURM time limit, hrs (default is 4hrs)

        -a <account> - SLURM account (default is your main account)

        -b - Use burst SLURM qos (default is main qos)

        -n - Do not wait until the job starts. I will run xpra_list_sessions later

        -j - Set up environment for a Java program

        -f <jobfile> - Job script to use for the gui session

        -l - List application presets

        -v - Verbose output to show the submission information


Defaults will be used for missing values

If you would like more detailed information on using advanced SLURM features with MATLAB gui sessions, please see the following job script examples.

Job Script Examples

See the Matlab_job_scripts page for matlab job script examples.

Additional Information

Compiled DLL

If you have compiled a DLL to use as an executable, it will not run on the cluster. DLL's are Microsoft Windows based dynamically linked libraries that will only work under the Windows operating system.

Setting a Command Path

To set a command path in your MatLab program, should add the following code.

path(path, '/path/to/matlab/binaries')

This will add the specified directory to the matlab search path. For more information see the Matlab web site.

Known Issues

No Licenses Available

You may run into an error when using the MatLab compiler such as

Licensed number of users already reached.

If so, the number of licenses available for the MatLab compiler has been exceeded. There are only two shared among the entire university so you should expect to see this problem from time to time if you use the MatLab compiler frequently. If the problem persists for more than a few hours, please open a support request.

MCR_CACHE_ROOT Directory

Could not access the MCR component cache

This error message means that MatLab is not able to access the MCR cache directory. When you load the MATLAB module, MCR_CACHE_ROOT is set for you, and you should not encounter this error. If you do, please open a support request. For batch jobs, module load matlab makes use of the $TMPDIR variable and MCR_CACHE_ROOT is set to $TMPDIR/mcr_cache. For non-batch jobs, MCR_CACHE_ROOT is set to /scratch/local/$USER/mcr_cache

Interactive Use

See the GUI section above for information on running the MATLAB GUI on HiPerGator.

Some Windows users running Xming have reported problems with MATLAB becoming unresponsive, especially after saving a new file. If you run into this problem, we suggest using an alternate X11 server, such as MobaXterm. Please open a support request if you continue to experience problems.

Older Versions

In order to use older versions of Matlab that are installed on the cluster, you can list them via the following command:

$ module spider matlab

If you were to see that an older version of matlab is available, such as version r2012a, you would then load the appropriate module for that version:

$ module load matlab/2012a

MDCS Testing

We have a Matlab MDCS trial version installed. Please download the MDCS configuration scripts for your local machine here: File:Matlab nonshared.tar.gz