[petsc-users] PETSc parallel scalability

[Jed Brown]

Jinlei Shen <jshen25 at jhu.edu> writes:

> Hi,
>
> Thanks a lot for the replies. They are really helpful.
>
> I just used the ksp ex2.c as an example to test the parallelization on my
> cluster, since the example have flexible number of unknowns.
> I do observe that for larger size of the problem, the speed-up shows better
> with more processors.

That example is intended to be readable by people new to PETSc, but it
is not a good choice for benchmarking because it uses a naive
distribution and solves an extremely easy problem.  Perhaps a large
deformation elasticity example (SNES ex16) would be more representative
of your target application?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 800 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20160908/a31b8d7a/attachment.pgp>