[petsc-users] Prometheus vs GAMG for elasticity/plasticity problems

[Thomas Gross]

I have now compiled Petsc 3.2.0 with Prometheus and Petsc 3.4.3 with GAMG using the same compiler. Furthermore I am using the same MPI implementation for both runs. Still Prometheus is considerably faster (see attached log files).

Prometheus:
-ksp_type cg -pc_type prometheus -log_summary -ksp_monitor -ksp_view -options_left > Prometheus_Large_Log_New.txt

Gamg:
-ksp_type cg -pc_type gamg -pc_gamg_type agg -pc_gamg_agg_nsmooths 1 -log_summary -ksp_monitor -ksp_view -options_left -mg_levels_ksp_max_it 1 > GAMG_Large_max_it_log_New.txt
-ksp_type cg -pc_type gamg -pc_gamg_type agg -pc_gamg_agg_nsmooths 1 -log_summary -ksp_monitor -ksp_view -options_left mg_levels_ksp_type richardson -mg_levels_pc_type sor > GAMG_Large_Rich_Log_New.txt

Best
Thomas


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Prometheus_Large_Log_New.txt
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140128/1d58a2d3/attachment-0003.txt>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: GAMG_Large_max_it_Log_New.txt
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140128/1d58a2d3/attachment-0004.txt>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: GAMG_Large_Rich_Log_New.txt
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140128/1d58a2d3/attachment-0005.txt>
-------------- next part --------------

Am 27.01.2014 um 21:38 schrieb Jed Brown <jed at jedbrown.org>:

> Thomas Gross <tgross at ilsb.tuwien.ac.at> writes:
> 
>> I was using a Newton Loop in Feap with a higher accuracy. This caused
>> FEAP8.4 to call KSPSolve with GAMG 3 times but FEAP8.3 to call
>> KSPSolve with Prometheus only 1 time.  I have now used a larger mesh
>> and adapted the FEAP solve file to call KSPSolve only once for GAMG
>> and Prometheus. However, Prometheus is about 3 times faster then GAMG
>> (see attached log files).
> 
> At the very least, you're using different compilers and a different MPI
> implementation (perhaps one is infiniband?).  Communication time is very
> different in these cases.  Should fix that before thinking harder about
> this.