[petsc-dev] Pipelined CG

[Jed Brown]

Yup, Matt is working with a student on exactly that, though in the context
of pipelined GMRES (which might be more important, the CG recurrence is
more sensitive to numerical stability).


On Wed, Dec 5, 2012 at 9:33 AM, Karl Rupp <rupp at mcs.anl.gov> wrote:

> Hi Jed,
>
> great, thanks! I think these versions might also work in a multi-GPU
> setting, where global synchronization is even more expensive. I should be
> able to give that a try soon :-)
>
> Best regards,
> Karli
>
>
>
>
> On 12/05/2012 11:08 AM, Jed Brown wrote:
>
>> I just merged several pipelined conjugate gradient methods from Pieter
>> Ghysels. If you are interested, here's the paper.
>>
>> http://www.exascience.com/wp-**content/uploads/2012/12/**
>> Ghysels_12.2012.1.pdf<http://www.exascience.com/wp-content/uploads/2012/12/Ghysels_12.2012.1.pdf>
>>
>> Access the methods with -ksp_type <pipecg,groppcg,pipecr>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20121205/7761dca2/attachment.html>