On 13 July 2011 21:06, Nevada Sanchez <[email protected]> wrote:
> I noticed that the 'multicore' package was running a little slow for large 
> data sets. I tracked the problem down to the 'save' -> 'load' interaction 
> between slaves and the master. It's pretty easy to fix. Just change
>
> save(resultFileName, 'result');
>
> to
>
> save('-binary', resultFileName, 'result');
>
> in 'startmulticoreslave.m' (line 103 in my source). Not sure why this wasn't 
> being done before, but I thought I'd pass it along.
>
> -Nevada

Hi Nevada,

That package hasn't been touched in 2 years. It's developer, Markus
Buehren, keeps its development in Mathworks website and has actually
updated it a few days ago
http://www.mathworks.com/matlabcentral/fileexchange/13775-multicore-parallel-processing-on-multiple-cores

The code is still under BSD license so in case his new version is no
longer compatible with octave. I wonder if his new version would still
be compatible with octave.

I checked the source of the new version and still doesn't have the
patch you mentioned.

Carnë

------------------------------------------------------------------------------
Magic Quadrant for Content-Aware Data Loss Prevention
Research study explores the data loss prevention market. Includes in-depth
analysis on the changes within the DLP market, and the criteria used to
evaluate the strengths and weaknesses of these DLP solutions.
http://www.accelacomm.com/jaw/sfnl/114/51385063/
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to