[Nek5000-users] Wall Clock Time
nek5000-users at lists.mcs.anl.gov
nek5000-users at lists.mcs.anl.gov
Thu Sep 4 13:09:11 CDT 2014
Hi,
What compiler and platform are you running on? Also what MPI
implementation are you using?
Thanks
Katherine
On Tue, Sep 2, 2014 at 4:19 PM, <nek5000-users at lists.mcs.anl.gov> wrote:
> Hello all,
>
> I'm interested in comparing the wall clock time for a same run with
> different number of CPU's. However, the time output seems to always give 0.
> I have tried using the mpi_wtime and dnekclock functions in the userchk
> subroutine and it has also been giving me 0 for each processor. I have
> included the end of my logfile in this email. The last 2 values of the
> first output line for a time step should give the total wall clock time and
> step wall clock time, but are both showing 0. At the end of the run, total
> elapsed time should be shown, but once again the value output is zero.
>
> Has anyone figured out a method permitting to obtain the total wall clock
> time?
>
> Thank you,
>
> Jérémie
>
> ----------------------- END OF LOGFILE ------------------------------
> ----------------------------------------------
>
>
> Step 20, t= 4.8774474E-02, DT= 2.4387237E-03, C= 0.501 0.0000E+00
> 0.0000E+00
> Solving for fluid
> 20 1 4.7138E-02 1.7111E-07 2.7549E+05 15 alph1x
> 20 halpha 15 -1.1778E-03 1.2672E-01 6.1694E-02 1.1588E-01
> 5.4007E-02 5.1022E-02 3.1403E-02 2.4425E-02 1.0726E-02 6.0344E-03
>
> 20 Hmholtz VELX: 3 2.6011E-13 1.7111E-07 1.0000E-12
> 20 2 3.4534E-02 2.5050E-07 1.3786E+05 15 alph1x
> 20 halpha 15 1.8538E-02 1.0186E-01 4.5783E-02 4.5495E-02
> 4.2007E-02 4.7957E-02 3.7328E-02 2.8192E-02 2.2477E-02 9.2507E-03
>
> 20 Hmholtz VELY: 3 3.4751E-13 2.5050E-07 1.0000E-12
> 20 15 alpha: -2.7079E-06 -1.8713E-04 3.2842E-03 3.7249E-03
> 2.1355E-03 1.4731E-03 1.7188E-03 3.0469E-03 1.6907E-03 1.1608E-03
> 20 15 4.0387E-03 7.8105E-07 5.1708E+03 alph12
> 20 U-PRES gmres: 13 8.2682E-13 1.0000E-12 1.0390E-09
> 0.0000E+00 0.0000E+00
> 20 DNORM, DIVEX 8.2682338981824994E-013
> 8.2682352659334355E-013
> 20 4.8774E-02 0.0000E+00 Fluid done
> call outfld: ifpsco: F
>
> 20 4.8774E-02 Write checkpoint:
> 0 20 OPEN: naca24150.f00001
>
> 20 4.8774E-02 done :: Write checkpoint
> file size = 4.4 MB
> avg data-throughput = InfMB/s
> io-nodes = 1
>
>
> end of time-step loop
>
> runtime statistics:
> total time 9.9999999999999995E-008
> inv3 time 621 0.0000000000000000 0.0000000000000000
> invc time 80 0.0000000000000000 0.0000000000000000
> mltd time 1282 0.0000000000000000 0.0000000000000000
> cdtp time 1282 0.0000000000000000 0.0000000000000000
> eslv time 20 0.0000000000000000 0.0000000000000000
> pres time 20 0.0000000000000000 0.0000000000000000
> crsl time 565 0.0000000000000000 0.0000000000000000
> crsl min 0.0000000000000000
> crsl max 0.0000000000000000
> crsl avg 0.0000000000000000
> hmhz time 8 0.0000000000000000 0.0000000000000000
> spro time 21 0.0000000000000000 0.0000000000000000
> usbc time 21 0.0000000000000000 0.0000000000000000
> usbc min 0.0000000000000000
> usbc max 0.0000000000000000
> usb avg 0.0000000000000000
> axhm time 458 0.0000000000000000 0.0000000000000000
> advc time 40 0.0000000000000000 0.0000000000000000
> vdss time 622 0.0000000000000000 0.0000000000000000
> vdss min 0.0000000000000000
> vdss max 0.0000000000000000
> vdss avg 0.0000000000000000
> dsum time 2139 0.0000000000000000 0.0000000000000000
> dsum min 0.0000000000000000
> dsum max 0.0000000000000000
> dsum avg 0.0000000000000000
> dadd time 0 0.0000000000000000 0.0000000000000000
> ddsl time 565 0.0000000000000000 0.0000000000000000
> solv time 565 0.0000000000000000 0.0000000000000000
> prep time 20 0.0000000000000000 0.0000000000000000
>
> # nid tusbc tdadd tcrsl tvdss tdsum tgop
> qqq F
> 0 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
> 0.0000E+00 qqq
> 1 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
> 0.0000E+00 qqq
> 2 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
> 0.0000E+00 qqq
> 3 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
> 0.0000E+00 qqq
> 4 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
> 0.0000E+00 qqq
> 5 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
> 0.0000E+00 qqq
>
> call exitt: dying ...
>
> backtrace(): obtained 5 stack frames.
> ./nek5000(print_stack_+0x1a) [0x4fa15a]
> ./nek5000(exitt_+0x1a6) [0x5e6b56]
> ./nek5000(main+0x39) [0x419489]
> /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xed) [0x2b7993c6076d]
> ./nek5000() [0x4194b9]
>
> total elapsed time : 0.00000E+00 sec
> total solver time incl. I/O : 0.00000E+00 sec
> time/timestep : 0.00000E+00 sec
> CPU seconds/timestep/gridpt : 0.00000E+00 sec
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20140904/3288ec65/attachment.html>
More information about the Nek5000-users
mailing list