[Cado-nfs-discuss] Using GPU in Sieving Step

Zimmermann Paul Paul.Zimmermann at loria.fr
Mon Aug 20 14:22:37 CEST 2012


       Hi Hamid,

> I tried GMP-ECM project (the branch that is currently under active
> development for GPU devices) on a GPU cluster and got good results.
> I want to replace ecm part of sieving step in CADO-NFS with GMP-ECM(that
> support GPU), but I have the following questions:
> 
> - Basically, it's a good idea or not?
> - Is the GMP-ECM faster than ecm part of CADO-NFS? and if so, why the
> developers don't do this replacement,yet?
> In addition, if all things is OK and it's a good idea, it would be pleasure
> if you give me some tips and points before doing that.

several people already answered, but as an author of GMP-ECM, I confirm it's
not a good idea, since the ECM code in CADO-NFS cofactorization is specialized
for one-word, 1.5-word, two-words, ... and for specific B1/B2 bounds, and thus
can be much faster than GMP-ECM. You'll find all details in Alexander Kruppa
PhD Thesis: http://tel.archives-ouvertes.fr/tel-00477005/en/

Of course we could use GPU devices for the cofactorization step, but that would
require a specialized library.

Paul Zimmermann



More information about the Cado-nfs-discuss mailing list