This page has not been updated yet. The page does not reflect the transition from PBS to Slurm.
# MATLAB
## Introduction
MATLAB is available in versions R2015a and R2015b. There are always two variants of the release:
* Non-commercial or so-called EDU variant, which can be used for common research and educational purposes.
* Commercial or so-called COM variant, which can used also for commercial activities. Commercial licenses are much more expensive, so usually the commercial license has only a subset of features compared to the available EDU license.
To load the latest version of MATLAB load the module:
MATLAB (an abbreviation of "MATrix LABoratory") is a proprietary multi-paradigm programming language
and numeric computing environment developed by MathWorks.
MATLAB allows matrix manipulations, plotting of functions and data, implementation of algorithms,
creation of user interfaces, and interfacing with programs written in other languages.
```console
$ml MATLAB
```
## Installed Versions
The EDU variant is marked as default. If you need other version or variant, load the particular version. To obtain the list of available versions, use:
Version 2021a is e-infra licence, without cluster licenses - only basic functionality.
## MATLAB GUI
If you need to use the MATLAB GUI to prepare your MATLAB programs, you can use MATLAB directly on the login nodes. However, for all computations, use MATLAB on the compute nodes via PBS Pro scheduler.
If you need to use the MATLAB GUI to prepare your MATLAB programs, you can use MATLAB directly on the login nodes.
However, for all computations, use MATLAB on the compute nodes via Slurm workload manager.
If you require the MATLAB GUI, follow the general information about [running graphical applications][1].
MATLAB GUI is quite slow using the X forwarding built in the PBS (`qsub -X`), so using X11 display redirection either via SSH or directly by `xauth` (see the [GUI Applications on Compute Nodes over VNC][1] section) is recommended.
MATLAB GUI is quite slow using the X forwarding built in the PBS (`qsub -X`),
so using X11 display redirection either via SSH or directly by `xauth`
(see the [GUI Applications on Compute Nodes over VNC][1] section) is recommended.
## Running Parallel MATLAB Using Distributed Computing Toolbox / Engine
## MATLAB Configuration
Distributed toolbox is available only for the EDU variant
### Client Configuration
The MPIEXEC mode available in previous versions is no longer available in MATLAB 2015. In addition, the programming interface has changed. Refer to [Release Notes][a].
After logging into the cluster, start MATLAB.
On the Home tab, click Parallel > Discover Clusters… to discover the profile.
Delete previously used file mpiLibConf.m, we have observed crashes when using Intel MPI.

To use Distributed Computing, you first need to setup a parallel profile. We have provided the profile for you, you can either import it in the MATLAB command line:
Jobs will now default to the cluster rather than submit to the local machine.
With the new mode, MATLAB itself launches the workers via PBS, so you can use either an interactive mode or a batch mode on one node, but the actual parallel processing will be done in a separate job started by MATLAB itself. Alternatively, you can use a "local" mode to run parallel code on just a single node.
The following example shows how to start the interactive session with support for MATLAB GUI. For more information about GUI based applications, see [this page][1].
>> % Specify the wall time (e.g., 1 day, 5 hours, 30 minutes)
>> c.AdditionalProperties.WallTime = '1-05:30';
```
This `qsub` command example shows how to run MATLAB on a single node.
Save changes after modifying AdditionalProperties for the above changes to persist between MATLAB sessions.
The second part of the command shows how to request all necessary licenses. In this case, 1 MATLAB-EDU license and 48 Distributed Computing Engines licenses.
```
>> c.saveProfile
```
Once the access to compute nodes is granted by PBS, the user can load following modules and start MATLAB:
To see the values of the current configuration options, display AdditionalProperties.
```console
$ml MATLAB/R2015b
$matlab &
```
>> % To view current properties
>> c.AdditionalProperties
```
### Parallel MATLAB Batch Job in Local Mode
Unset a value when no longer needed.
To run MATLAB in a batch mode, write a MATLAB script, then write a bash jobscript and execute via the `qsub` command. By default, MATLAB will execute one MATLAB worker instance per allocated core.
To run an interactive pool job on the cluster, continue to use `parpool` as before.
# load modules
ml MATLAB/R2015b
```
>> % Get a handle to the cluster
>> c = parcluster;
# execute the calculation
matlab -nodisplay-r matlabcode > output.out
>> % Open a pool of 64 workers on the cluster
>> pool = c.parpool(64);
```
# copy output file to home
cp output.out $PBS_O_WORKDIR/.
Rather than running local on the local machine, the pool can now run across multiple nodes on the cluster.
# remove scratch folder
rm-rf$SCR
```
>> % Run a parfor over 1000 iterations
>> parfor idx = 1:1000
a(idx) = rand;
end
```
# exit
exit
Delete the pool when it’s no longer needed.
```
>> % Delete the pool
>> pool.delete
```
This script may be submitted directly to the PBS workload manager via the `qsub` command. The inputs and the MATLAB script are in the matlabcode.m file, outputs in the output.out file. Note the missing .m extension in the `matlab -r matlabcodefile` call, **the .m must not be included**. Note that the **shared /scratch must be used**. Further, it is **important to include the `quit`** statement at the end of the matlabcode.m script.
### Independent Batch Job
Submit the jobscript using `qsub`:
Use the batch command to submit asynchronous jobs to the cluster.
The batch command will return a job object which is used to access the output of the submitted job.
See the MATLAB documentation for more help on batch.
```console
$qsub ./jobscript
```
>> % Get a handle to the cluster
>> c = parcluster;
### Parallel Matlab Local Mode Program Example
>> % Submit job to query where MATLAB is running on the cluster
>> job = c.batch(@pwd, 1, {});
The last part of the configuration is done directly in the user's MATLAB script before Distributed Computing Toolbox is started.
>> % Query job for state
>> job.State
```console
cluster = parcluster('local')
>> % If state is finished, fetch the results
>> job.fetchOutputs{:}
>> % Delete the job after results are no longer needed
>> job.delete
```
This script creates the scheduler object *cluster* of the type *local* that starts workers locally.
To retrieve a list of running or completed jobs, call `parcluster` to return the cluster object.
The cluster object stores an array of jobs that are queued to run, are running, have run, or have failed.
Retrieve and view the list of jobs as shown below.
!!! hint
Every MATLAB script that needs to initialize/use `matlabpool` has to contain these three lines prior to calling the `parpool(sched, ...)` function.
```
>> c = parcluster;
>> jobs = c.Jobs
>>
>> % Get a handle to the second job in the list
>> job2 = c.Jobs(2);
```
The last step is to start `matlabpool` with the *cluster* object and a correct number of workers. We have 128 cores per node, so we start 128 workers.
Once the job has been selected, fetch the results as previously done.
```console
parpool(cluster,128);
`fetchOutputs` is used to retrieve function output arguments; if calling `batch` with a script, use `load` instead.
Data that has been written to files on the cluster needs be retrieved directly from the file system (e.g., via sftp).
```
>> % Fetch all results from the second job in the list
>> job2.fetchOutputs{:}
```
... parallel code ...
### Parallel Batch Job
`batch` can also submit parallel workflows.
Let’s use the following example for a parallel job, which is saved as `parallel_example.m`.
parpool close
```
function [sim_t, A] = parallel_example(iter)
The complete example showing how to use Distributed Computing Toolbox in local mode is shown here.
if nargin==0
iter = 8;
end
```matlab
cluster=parcluster('local');
cluster
disp('Start sim')
parpool(cluster,128);
t0 = tic;
parfor idx = 1:iter
A(idx) = idx;
pause(2)
idx
end
sim_t = toc(t0);
n=2000;
disp('Sim completed')
W=rand(n,n);
W=distributed(W);
x=(1:n)';
x=distributed(x);
spmd
[~,name]=system('hostname')
save RESULTS A
T=W*x;% Calculation performed on labs, in parallel.
% T and W are both codistributed arrays here.
end
T;
whos% T and W are both distributed arrays here.
```
This time when using the `batch` command, also specify a MATLAB `Pool` argument.
delete(gcp('nocreate'))% close parpool
quit
```
>> % Get a handle to the cluster
>> c = parcluster;
You can copy and paste the example in a .m file and execute. Note that the `parpool` size should correspond to the **total number of cores** available on allocated nodes.
>> % Submit a batch pool job using 4 workers for 16 simulations
### Parallel MATLAB Batch Job Using PBS Mode (Workers Spawned in a Separate Job)
>> % View current job status
>> job.State
This mode uses the PBS scheduler to launch the parallel pool. It uses the KarolinaPBSPro profile that needs to be imported to Cluster Manager, as mentioned before. This method uses MATLAB's PBS Scheduler interface - it spawns the workers in a separate job submitted by MATLAB using qsub.
>> % Fetch the results after a finished state is retrieved
>> job.fetchOutputs{:}
This is an example of an m-script using the PBS mode:
>> % Clear job from workspace (as though MATLAB exited)
>> clear job
```
T=W*x;% Calculation performed on labs, in parallel.
% T and W are both codistributed arrays here.
end
whos% T and W are both distributed arrays here.
With a handle to the cluster, the `findJob` method searches for the job with the specified job ID.
% shut down parallel pool
delete(gcp('nocreate'))
quit
```
>> % Get a handle to the cluster
>> c = parcluster;
Note that we first construct a cluster object using the imported profile, then set some important options, namely: `SubmitArguments`, where you need to specify accounting id, and `ResourceTemplate`, where you need to specify the number of nodes to run the job.
>> % Find the old job
>> job = c.findJob('ID', 4);
You can start this script using the batch mode the same way as in the Local mode example.
>> % Retrieve the state of the job
>> job.State
### Non-Interactive Session and Licenses
ans =
finished
If you want to run batch jobs with MATLAB, be sure to request appropriate license features with the PBS Pro scheduler, at least the `-l license__matlab-edu__MATLAB=1` for the EDU variant of MATLAB. For more information about how to check the license features states and how to request them with PBS Pro, [look here][3].
>> % Fetch the results
>> job.fetchOutputs{:};
In case of non-interactive session, read the [following information][3] on how to modify the `qsub` command to test for available licenses prior getting the resource allocation.