<div dir="auto">Thank you Barry, sounds great. I'll try it out in the next weeks! Is copy data such a bottleneck with respect to the solving time in your opinion? I am not scared of building stuff on gpu directly, I basically assemble the petsc matrix and rhs in one point so it would be ok doing stuff on gpu directly. Is aij format ok for gpu or better CSR?<div dir="auto"><br></div><div dir="auto">Cheers.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Il Dom 25 Ago 2024, 13:49 Edoardo alinovi <<a href="mailto:edoardo.alinovi@gmail.com">edoardo.alinovi@gmail.com</a>> ha scritto:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto">Hello Petsc friends,<div dir="auto"><br></div><div dir="auto">As many people is doing, I would like to explore a bit gpu capabilities (cuda) in petsc. </div><div dir="auto"><br></div><div dir="auto">Before attemping any coding effort I would like to hear from you if all of this make sense:</div><div dir="auto">- compile mpi with cuda support</div><div dir="auto">- compile petsc with cuda support</div><div dir="auto">- build matrix and vectors as MATAIJCUSPARSE and VECMPICUDA to tell petsc using gpu.</div><div dir="auto"><br></div><div dir="auto">That's really it or do I need to take care of something else? </div><div dir="auto"><br></div><div dir="auto">I have seen that there is an amgXWrapper library aroud, but not sure if it is still relevant now or not.</div><div dir="auto"><br></div><div dir="auto">Thank you for the suggestions!</div><div dir="auto"><br></div><div dir="auto"><br></div></div>
</blockquote></div>