[SciPy-dev] Building newcore with the intel compilers on Itanium2...

[Fernando Perez]

Hey Arnd,

Arnd Baecker wrote:

> Just out of curiosity: how big are the benefits of icc?
> (The computing center here in Dresden is just
> installing the first parts of 1500 ItaniumR2
> and so we will very much benifit from your pioneering work on
> Itanium ...).
> Do you also have the Intel Math Kernel Library
> http://www.intel.com/cd/software/products/asmo-na/eng/perflib/mkl/index.htm
> ?
> It would be very interesting to see how much performance
> increase one can get in practice when using their
> Linear Algebra, FFT, vector math, ...
> compared to ATLAS, fftw, etc.).
> For example, for fft it seems to be really good:
> http://www.intel.com/cd/software/products/asmo-na/eng/perflib/mkl/219662.htm
> And also for atlas (especially multiprocesssor support):
> http://www.intel.com/cd/software/products/asmo-na/eng/perflib/mkl/219823.htm
> 
> ((No - I don't have stock options with them ;-))

Well, I was really just helping Andrew (a colleague and part of my 'convert or 
exterminate the matlab users' crusade ;), so I'll let him answer in more 
detail.  On my own codes, a while back I did some tests with icc/ifort and 
didn't notice improvements that were worthwhile enough to deal with the 
hassles.  Going with gcc is just the path of least resistance, especially back 
in the intel 7.0 days which had far less gcc compatibility than they do now.

> ((Remark: even if all the unit tests pass, it does
> not mean that code which used Numeric directly works with newcore
> - we just had such a case yesterday, where it should have worked ...
> so far we have not yet managed to isolate the problem.))

Good point.  I haven't converted real code yet, so far just testing builds and 
so forth.

Cheers,

f