[Users] Einstein Toolkit Meeting Reminder

Mon Jun 7 14:30:45 CDT 2021

Bruno

I don't have any official files to share. I'll attach my own
configuration files which depend on many self-installed software
packages.

If you are interested in using GPUs, then we can meet on Zoom (with
your collaborators or students, and others who are interested), and I
can give a brief demonstration.

-erik

On Thu, Jun 3, 2021 at 10:39 AM Bruno Giacomazzo
<bruno.giacomazzo at unimib.it> wrote:
>
> Erik,
>   thanks. Do you have any instructions (or simfactory files) to share if I wanted to try compiling and running it on GPUs?
> I have access to a GPU cluster at CINECA in Italy and I may try running there (each node has 4 x NVIDIA Volta V100 GPUs). The theory group here in Bicocca is also going to buy a server with 4 x NVIDIA A100 GPUs for testing purposes.
>
> I do not have experience with GPUs, but I'm happy to help if needed.
>
> Cheers,
> Bruno
>
>
>
> Il giorno gio 3 giu 2021 alle ore 15:38 Erik Schnetter <schnetter at cct.lsu.edu> ha scritto:
>>
>> Bruno
>>
>> GPUs are supported and are working fine. A WaveToy is working, but I
>> haven't tested it in the past weeks. The Einstein equations require
>> more work to be efficient on GPUs.
>>
>> -erik
>>
>> On Thu, Jun 3, 2021 at 9:29 AM Bruno Giacomazzo
>> <bruno.giacomazzo at unimib.it> wrote:
>> >
>> > Roland and Erik,
>> >   thanks a lot for the explanation. I saw the release notes with CUDA listed in the larger changes and this is why I asked.
>> >
>> > Erik, does CarpetX already support GPU via AMReX or is more work needed?
>> >
>> > Thanks,
>> > Bruno
>> >
>> >
>> > Il giorno gio 3 giu 2021 alle ore 14:35 Erik Schnetter <schnetter at cct.lsu.edu> ha scritto:
>> >>
>> >> To my knowledge, the improved support means that it is now possible to
>> >> mix CUDA and C++ code in the same source files. That is, you can
>> >> compile C++ code with a CUDA compiler, which will pass on the C++ bits
>> >> to the regular C++ compiler while extracting the CUDA code. This is
>> >> merely a technical improvement of the make system that makes it easier
>> >> to write one version code that can run on both a CPU or a GPU, instead
>> >> of having to write two separate versions of the code.
>> >>
>> >> The real CUDA-related work happened in CarpetX
>> >> <https://bitbucket.org/eschnett/cactusamrex/>, which is not part of
>> >> the Einstein Toolkit yet. But we're getting there!
>> >>
>> >> -erik
>> >>
>> >> On Thu, Jun 3, 2021 at 3:40 AM Bruno Giacomazzo
>> >> <bruno.giacomazzo at unimib.it> wrote:
>> >> >
>> >> > would it be possible to have a (short) summary of the CUDA support introduced in the latest release?
>> >> >
>> >> > Thanks,
>> >> > Bruno
>> >> >
>> >> > Il giorno gio 3 giu 2021 alle ore 00:15 <users at einsteintoolkit.org> ha scritto:
>> >> >>
>> >> >> Hello,
>> >> >>
>> >> >> Please consider joining the weekly Einstein Toolkit phone call at
>> >> >> 9:00 am US central time on Thursdays. For details on how to connect
>> >> >> and what agenda items are to be discussed, use the link below.
>> >> >>
>> >> >> https://docs.einsteintoolkit.org/et-docs/Main_Page#Weekly_Users_Call
>> >> >>
>> >> >> --The Maintainers
>> >> >> _______________________________________________
>> >> >> Users mailing list
>> >> >> Users at einsteintoolkit.org
>> >> >> http://lists.einsteintoolkit.org/mailman/listinfo/users
>> >> >
>> >> >
>> >> >
>> >> > --
>> >> >
>> >> > Prof. Bruno Giacomazzo
>> >> > Department of Physics
>> >> > University of Milano-Bicocca
>> >> > Piazza della Scienza 3
>> >> > 20126 Milano
>> >> > Italy
>> >> >
>> >> > email: bruno.giacomazzo at unimib.it
>> >> > phone: (+39) 02 6448 2321
>> >> > web: http://www.brunogiacomazzo.org
>> >> >
>> >> > ----------------------------------------------------------------------
>> >> > There are only 10 types of people in the world:
>> >> > Those who understand binary, and those who don't
>> >> > ----------------------------------------------------------------------
>> >> >
>> >> > _______________________________________________
>> >> > Users mailing list
>> >> > Users at einsteintoolkit.org
>> >> > http://lists.einsteintoolkit.org/mailman/listinfo/users
>> >>
>> >>
>> >>
>> >> --
>> >> Erik Schnetter <schnetter at cct.lsu.edu>
>> >> http://www.perimeterinstitute.ca/personal/eschnetter/
>> >
>> >
>> >
>> > --
>> >
>> > Prof. Bruno Giacomazzo
>> > Department of Physics
>> > University of Milano-Bicocca
>> > Piazza della Scienza 3
>> > 20126 Milano
>> > Italy
>> >
>> > email: bruno.giacomazzo at unimib.it
>> > phone: (+39) 02 6448 2321
>> > web: http://www.brunogiacomazzo.org
>> >
>> > ----------------------------------------------------------------------
>> > There are only 10 types of people in the world:
>> > Those who understand binary, and those who don't
>> > ----------------------------------------------------------------------
>> >
>> > _______________________________________________
>> > Users mailing list
>> > Users at einsteintoolkit.org
>> > http://lists.einsteintoolkit.org/mailman/listinfo/users
>>
>>
>>
>> --
>> Erik Schnetter <schnetter at cct.lsu.edu>
>> http://www.perimeterinstitute.ca/personal/eschnetter/
>
>
>
> --
>
> Prof. Bruno Giacomazzo
> Department of Physics
> University of Milano-Bicocca
> Piazza della Scienza 3
> 20126 Milano
> Italy
>
> email: bruno.giacomazzo at unimib.it
> phone: (+39) 02 6448 2321
> web: http://www.brunogiacomazzo.org
>
> ----------------------------------------------------------------------
> There are only 10 types of people in the world:
> Those who understand binary, and those who don't
> ----------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at einsteintoolkit.org
> http://lists.einsteintoolkit.org/mailman/listinfo/users

-- 
Erik Schnetter <schnetter at cct.lsu.edu>
http://www.perimeterinstitute.ca/personal/eschnetter/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: symmetry-gpu.cfg
Type: application/octet-stream
Size: 4325 bytes
Desc: not available
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20210607/49e1222a/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: symmetry-gpu.sub
Type: application/octet-stream
Size: 530 bytes
Desc: not available
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20210607/49e1222a/attachment-0001.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: symmetry-gpu.run
Type: application/octet-stream
Size: 4775 bytes
Desc: not available
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20210607/49e1222a/attachment-0002.obj 