[petsc-users] GPU speedup in Poisson solvers

[Jed Brown]

Dominic Meiser <dmeiser at txcorp.com> writes:
> - To get reliable timing you should configure PETSc without debugging 
> (i.e. --with-debugging=no)
> - The ILU preconditioning in your GPU benchmark is done on the CPU. The 
> host-device data transfers are killing performance. Can you try to run 
> with the additional option --pc_factor_mat_solver_packe cusparse? This 
> will perform the preconditioning on the GPU.
> - If you're interested in running benchmarks in parallel you will need a 
> few patches that are not yet in petsc/master. I can put together a 
> branch that has the needed fixes.

And for the CPU version, considering using a configuration that makes
sense there.  Like FMG with Gauss-Seidel or Chebyshev smoothers and an
error tolerance proportional to discretization error.  You might find
that not enough time is spent on the fine grid to see a significant
speed-up.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140922/cddd498c/attachment.pgp>