Hi Renat,
MATLAB is a single-threaded/process application. Left to its own devices it
will only use one processor regardless of how many processors/nodes you have.
There are a couple solutions with regards to MATLAB in general, and SPM in
particular. Firstly for MATLAB:
1. Mathworks has released a Distributed Computing Toolbox:
http://www.mathworks.com/products/distribtb/
2. The MATLAB user community has several home-brewed solutions which can be
found on the Mathworks Matlabcentral file exchange:
http://www.mathworks.com/matlabcentral/fileexchange/loadCategory.do?objectType=category&objectId=111&objectName=Distributed%20Processing
With regards to SPM:
1. there is a parallelization toolbox available for SPM2 which will parallelize
realignment, slice-timing, normalization, smoothing, and statistics:
http://sourceforge.net/projects/parallelspm
2. There are other toolboxes available for SPM99 (which I am unfamiliar with),
which can be googled
3. The SPM Wiki might give you more information:
http://en.wikibooks.org/wiki/SPM-Faster_SPM#Clusters_and_parallel_processing
Regards,
Jejo
Quoting Renat Yakupov <[log in to unmask]>:
> Hello everybody.
>
> Does anybody have experience running Matlab on a Linux cluster? Is it
> possible at all? Is Matlab a, what it is called, cluster-aware application?
> What about running it on a multi-processor workststation? Up to how many can
> Matlab handle?
> Thank you.
>
> Renat.
>
>
|