[CIG-SEISMO] Increasing specfem Grid

[Dimitri Komatitsch]

Dear Leandro,

Carene is right, you need to divide the time step by 256 / 128 = 2 
(since you switch to a twice-denser grid, in an explicit time stepping 
scheme you need to divide the time step by two to remain below the CFL 
stability limit of the time scheme).

(this is a general property of explicit time schemes, not specifically 
related to SPECFEM)

Best wishes,
Dimitri.

On 1/7/19 10:24 PM, Larmat, Carene wrote:
> Did you change your DT accordingly (page 36 of the manual)?
> 
> Carene Larmat, EES-17
> LANL, carene at lanl.gov <mailto:carene at lanl.gov>, 505 667 2074
> 
> 
> 
>> On Jan 7, 2019, at 11:42 AM, Leandro Gazoni <lgazoni at gmail.com 
>> <mailto:lgazoni at gmail.com>> wrote:
>>
>>
>> Dear,
>>
>> My name is Leandro Gazoni, I have a doctorate in computational 
>> mechanics from the Federal University of Rio de Janeiro / Brazil. I'm 
>> doing some studies with the use of SprecFem3d. I'm currently having 
>> problems running specfem on a cluster.
>>
>> I have a simple case that I can run with no errors in the clause with 
>> a grid size of 128x128x30. When I try to refine the gride of this same 
>> problem to the dimensions of 256x256x60. Specfem displays the 
>> following error. Please can you help me.
>>
>> ---
>> File: T02-256x256x60-16p.o137647
>>
>> Program received signal SIGFPE: Floating-point exception - erroneous 
>> arithmetic operation.
>>
>> Backtrace for this error:
>> --------------------------------------------------------------------------
>> A process has executed an operation involving a call to the
>> "fork()" system call to create a child process.  Open MPI is currently
>> operating in a condition that could result in memory corruption or
>> other system errors; your job may hang, crash, or produce silent
>> data corruption.  The use of fork() (or system() or other calls that
>> create child processes) is strongly discouraged.
>>
>> The process that invoked fork was:
>>
>>   Local host:          [[26656,1],5] (PID 20263)
>>
>> If you are *absolutely sure* that your application will successfully
>> and correctly survive a call to fork(), you may disable this warning
>> by setting the mpi_warn_on_fork MCA parameter to 0.
>> --------------------------------------------------------------------------
>> #0  0x2AAAAB448507
>> #1  0x2AAAAB447700
>> #2  0x2AAAAC0F613F
>> #3  0x42A5D1 in compute_forces_viscoelastic_ at 
>> compute_forces_viscoelastic.f90:630
>> #4  0x424507 in compute_forces_viscoelastic_calling_ at 
>> compute_forces_viscoelastic_calling_routine.F90:71
>> #5  0x48D0D5 in iterate_time_ at iterate_time.F90:241 (discriminator 1)
>>
>> Program received signal SIGFPE: Floating-point exception - erroneous 
>> arithmetic operation.
>>
>> Backtrace for this error:
>>
>> Program received signal SIGFPE: Floating-point exception - erroneous 
>> arithmetic operation.
>>
>> Backtrace for this error:
>>
>> Program received signal SIGFPE: Floating-point exception - erroneous 
>> arithmetic operation.
>>
>> Backtrace for this error:
>> --------------------------------------------------------------------------
>> mpirun noticed that process rank 5 with PID 0 on node r2i1n11 exited 
>> on signal 8 (Floating point exception).
>> --------------------------------------------------------------------------
>> [r2i1n11:20253] 3 more processes have sent help message 
>> help-opal-runtime.txt / opal_init:warn-fork
>> [r2i1n11:20253] Set MCA parameter "orte_base_help_aggregate" to 0 to 
>> see all help / error messages
>> ---
>>
>>
>> Regards,
>> Leandro Gazoni
>> _________________________________________________
>> CIG-SEISMO has MOVED to our new forum.  Please
>> head to https://community.geodynamics.org.
>>
>> This mailing list will be disabled at the end of
>> January 2019.
>> _________________________________________________
>> CIG-SEISMO mailing list
>> CIG-SEISMO at geodynamics.org <mailto:CIG-SEISMO at geodynamics.org>
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo
> 
> 
> _________________________________________________
> CIG-SEISMO has MOVED to our new forum.  Please
> head to https://community.geodynamics.org.
> 
> This mailing list will be disabled at the end of
> January 2019.
> _________________________________________________
> CIG-SEISMO mailing list
> CIG-SEISMO at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo
> 

-- 
Dimitri Komatitsch, CNRS Research Director (DR CNRS)
Laboratory of Mechanics and Acoustics, Marseille, France
http://komatitsch.free.fr