Ya thanks for the suggestion. strangely it worked. However, if I had not
included hypre, the original command also worked.
Tks anyway.
On 2/12/07, Satish Balay balay at mcs.anl.gov wrote:
- If you have build isses [involing sending configure.log] please use
petsc-maint at mcs.anl.gov address
Well,
I am simulating unsteady flow past a moving airfoil at Re~10^4. I'm using
fractional step FVM, which means that I need to solve a momentum and poisson
equation.
To reach a periodic state takes quite a few hours and so I'm trying to find
ways to speed up the process. I thought parallelizing
- If you have build isses [involing sending configure.log] please use
petsc-maint at mcs.anl.gov address [not the mailing list]
- Looks like you were using the following configure options:
--with-cc=/scratch/g0306332/intel/cc/bin/icc
--with-fc=/lsftmp/g0306332/inter/fc/bin/ifort
Ya, that's the mistake. I changed part of the code resulting in
PetscFinalize not being called.
Here's the output:
-- PETSc Performance Summary:
--
/home/enduser/g0306332/ns2d/a.out on a linux-mpi named
Hi,
I've repeated the test with n,m = 800. Now serial takes around 11mins while
parallel with 4 processors took 6mins. Does it mean that the problem must be
pretty large before it is more superior to use parallel? Moreover 800x800
means there's 64 unknowns. My problem is a 2D CFD code which
My recommendation is just to try to optimize sequential runs
by using the most appropriate solver algorithms, the best sequential
processor with the fastest memory and slickest code.
Parallel computing is to solve big problems, not to solve little problems
fast. (anything less then 100k
Hi,
I've tried to use log_summary but nothing came out? Did I miss out
something? It worked when I used -info...
On 2/9/07, Lisandro Dalcin dalcinl at gmail.com wrote:
On 2/8/07, Ben Tay zonexo at gmail.com wrote:
i'm trying to solve my cfd code using PETSc in parallel. Besides the
linear
-log_summary
On Fri, 9 Feb 2007, Ben Tay wrote:
Hi,
I've tried to use log_summary but nothing came out? Did I miss out
something? It worked when I used -info...
On 2/9/07, Lisandro Dalcin dalcinl at gmail.com wrote:
On 2/8/07, Ben Tay zonexo at gmail.com wrote:
i'm trying to
Impossible, please check the spelling, and make sure your
command line was not truncated.
Matt
On 2/9/07, Ben Tay zonexo at gmail.com wrote:
ya, i did use -log_summary. but no output.
On 2/9/07, Barry Smith bsmith at mcs.anl.gov wrote:
-log_summary
On Fri, 9 Feb 2007, Ben
Well, I don't know what's wrong. I did the same thing for -info and it
worked. Anyway, is there any other way?
Like I can use -mat_view or call matview( ... ) to view a matrix. Is there a
similar subroutine for me to call?
Thank you.
On 2/9/07, Matthew Knepley knepley at gmail.com wrote:
Problems do not go away by ignoring them. Something is wrong here, and it
may
affect the rest of your program. Please try to run an example:
cd src/ksp/ksp/examples/tutorials
make ex2
./ex2 -log_summary
Matt
On 2/9/07, Ben Tay zonexo at gmail.com wrote:
Well, I don't know what's
On 2/9/07, Ben Tay zonexo at gmail.com wrote:
ops it worked for ex2 and ex2f ;-)
so what could be wrong? is there some commands or subroutine which i must
call? btw, i'm programming in fortran.
Yes, you must call PetscFinalize() in your code.
Matt
thank you.
On 2/9/07, Matthew
1) These MFlop rates are terrible. It seems like your problem is way too
small.
2) The load balance is not good.
Matt
On 2/9/07, Ben Tay zonexo at gmail.com wrote:
Ya, that's the mistake. I changed part of the code resulting in
PetscFinalize not being called.
Here's the output:
Hi,
i'm trying to solve my cfd code using PETSc in parallel. Besides the linear
eqns for PETSc, other parts of the code has also been parallelized using
MPI.
however i find that the parallel version of the code running on 4 processors
is even slower than the sequential version.
in order to find
On 2/8/07, Ben Tay zonexo at gmail.com wrote:
i'm trying to solve my cfd code using PETSc in parallel. Besides the linear
eqns for PETSc, other parts of the code has also been parallelized using
MPI.
Finite elements or finite differences, or what?
however i find that the parallel version of
15 matches
Mail list logo
