[Clp] skylake xeons and AVX-512

jajhall at ed.ac.uk jajhall at ed.ac.uk
Sat Feb 25 11:11:33 EST 2017


Dear All,

Indeed, for LP problems where solution time is dominated by the matrix-vector product in PRICE there is some scope for vectorization. This was a (disused) feature of CPLEX 10-20 years ago.

However, there is still minimal computation for each memory access so, particularly with modern architectures, performance improvement is limited by the number of memory channels rather than number of cores.

For general LP problems the amount of such task parallelism isn't enough to get meaningful overall performance gain.

Julian

On 25 February 2017 15:53:46 GMT+00:00, John Forrest <john.forrest at fastercoin.com> wrote:
>David,
>
>I have done a bit of work to use AVX2 on Haswell architecture.  At 
>present it is only in pricing out matrix as that seemed the best 
>candidate for vector instructions.  I was partly just playing around
>and 
>the code was aimed at primal simplex.  For dense factorizations, I am 
>sure openblas etc will be upgraded.
>
>There were alignment issues on Haswell, and I can look to see what is 
>new on later architectures.
>
>The idea was to be able to generate a copy of matrix suitable for
>vector 
>instructions.  This was done using some dubious parameters and -DAVX2=4
>
>(for Haswell) in configure to get performance improvement.
>
>John Forrest
>On 25/02/17 11:04, David Prime wrote:
>> Hi all,
>>
>> With the release of the new Xeon Skylake CPUs imminent and google 
>> compute cloud offering early access to servers running them, I'm 
>> wondering if there's any pre existing work or thoughts on how CLP 
>> performs with these extra SIMD extensions?
>>
>> I'm planning to get hold of a few servers as soon as I can and
>compile 
>> the lib with various compilers/settings and see if there are any nice
>
>> automatic optimisations to be had. I'll share any benchmarks I 
>> generate and maybe we can see if there's any further room for 
>> optimisation on these new architectures.
>>
>> Are there any clp specific build flags to enable/disable certain 
>> features that could be relevant here?
>>
>> Cheers,
>> David
>>
>>
>> _______________________________________________
>> Clp mailing list
>> Clp at list.coin-or.org
>>
>https://urldefense.proofpoint.com/v2/url?u=http-3A__list.coin-2Dor.org_mailman_listinfo_clp&d=DwIC-g&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=JO30qYlZJBnOYqNHaz_LLqRwIFx89cT9BYhx1vmS6QM&m=v7KRvkPyYVoa403NU9HsD4-3sxIvf93OSIVk60rCtrU&s=KxM0suqn5aHDkOaqzKC8iAJUOlSj1xr5HfBbAmPwt2E&e=
>

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/clp/attachments/20170225/3fcdedb8/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <http://list.coin-or.org/pipermail/clp/attachments/20170225/3fcdedb8/attachment.ksh>


More information about the Clp mailing list