Re: octave + mkl

[Sergei Steshenko]

----- Original Message -----
> From: Jordi Gutiérrez Hermoso <address@hidden>
> To: Joshua Dillon <address@hidden>
> Cc: Sergei Steshenko <address@hidden>; "address@hidden" <address@hidden>
> Sent: Wednesday, May 16, 2012 11:37 PM
> Subject: Re: octave + mkl
> 
> On 16 May 2012 16:29, Joshua Dillon <address@hidden> wrote:
> 
>>  On Wed, May 16, 2012 at 1:28 PM, Joshua Dillon <address@hidden>
>>  wrote:
>>> 
>>>  My *VERY* informal timing of MKL+octave shows it to be exactly the
>>>  same speed as Matlab for "standard" operations like eig and 
> eigs.
>>>  Indeed, matlab uses MKL for these operations.
> 
> eigs has little to do with MKL. It's done with ARPACK. Are you really
> seeing a lot of difference with eigs?
> 
>>  Also--I'm on a quad core i5 2405 so that probably explains the 4x
>>  speed-up.
> 
> So probably what we'd like to see is more parallelisation in OpenBLAS,
> LAPACK, and fftw?
> 
> - Jordi G. H.
>

People often want something (e.g. speed) _now_.

Regards,
  Sergei.