[Numpy-discussion] dynamically choosing atlas libraries

[David Verelst]

Hi,

Not a developer here, but I was under the impression that you can only use
the BLAS/LAPACK libraries that where chosen at build time?

As a side note: I've read [1] that OpenBLAS on some systems could perform
quite well compared to ATLAS, I used some simple benchmarks [2] and noticed
that on my system (Intel Core2 Duo P9700) there was also a big performance
gain. I build standard LAPACK against OpenBLAS. OpenBlas has been mentioned
a few times in the last months on this list too.

[1]
http://www.der-schnorz.de/2012/06/optimized-linear-algebra-and-numpyscipy/
[2]
https://dpinte.wordpress.com/2010/01/15/numpy-performance-improvement-with-the-mkl/


On 5 April 2013 19:08, Edward Walter <ewalter+numpy at cs.cmu.edu> wrote:

> Hello List,
>
> We're standing up a new computational cluster and going through the
> steps of building Numpy etc.  We would like to be able to install
> multiple versions of ATLAS (with different build settings / tunings /
> etc) and have Numpy load the shared Atlas libraries dynamically based on
> the LD_LIBRARY_PATH.
>
> It's not clear to me how to do this given that the library path is hard
> coded in Numpy's site.cfg.  It seems like other people have gotten this
> working though (i.e. RHEL/Centos with their collection of Atlas versions
> {atlas, atlas-sse3, etc}).  Does anyone have any pointers on this?
>
> Thanks much.
>
> -Ed Walter
> Carnegie Mellon University
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion at scipy.org
> http://mail.scipy.org/mailman/listinfo/numpy-discussion
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/numpy-discussion/attachments/20130407/7629454f/attachment.html>