Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

Long simulation time on some of the time-steps

This post was from a previous version of the WRF&MPAS-A Support Forum. New replies have been disabled and if you have follow up questions related to this post, then please start a new thread from the forum home page.

Chang_Shu

New member
Hello, WRF team

I am running a 1km resolution WRF run with 3 nested domains for a future time frame.
I found that the simulation on some of the time-steps would take much longer time than most of the time-steps.
Please find an example snippet below I found from the rsl.* files:


Timing for main: time 2075-05-02_10:00:15 on domain 3: 0.04221 elapsed seconds
Timing for main: time 2075-05-02_10:00:15 on domain 2: 41.35098 elapsed seconds
Timing for main: time 2075-05-02_10:00:20 on domain 3: 0.03977 elapsed seconds
Timing for main: time 2075-05-02_10:00:25 on domain 3: 0.04215 elapsed seconds
Timing for main: time 2075-05-02_10:00:30 on domain 3: 0.03837 elapsed seconds
Timing for main: time 2075-05-02_10:00:30 on domain 2: 0.26364 elapsed seconds
Timing for main: time 2075-05-02_10:00:35 on domain 3: 0.03771 elapsed seconds
Timing for main: time 2075-05-02_10:00:40 on domain 3: 0.03286 elapsed seconds
Timing for main: time 2075-05-02_10:00:45 on domain 3: 0.04725 elapsed seconds
Timing for main: time 2075-05-02_10:00:45 on domain 2: 0.23428 elapsed seconds
Timing for main: time 2075-05-02_10:00:45 on domain 1: 57.36857 elapsed seconds
Timing for main: time 2075-05-02_10:00:50 on domain 3: 0.03843 elapsed seconds
Timing for main: time 2075-05-02_10:00:55 on domain 3: 0.04312 elapsed seconds
Timing for main: time 2075-05-02_10:01:00 on domain 3: 0.08528 elapsed seconds


This seriously slowed down my progress in the simulation.
Could you please let me know what would be the possible reason for this?

Thanks,

Chang
 
Hi Chang,
radiation scheme usually takes quite a long time. Can you look at your namelist.input and see whether raft is set to 30?
 
Hi, Ming
I am using three nested domains with spatial resolutions of 9km, 3km and 1km
My settings of radiation in my namelist.input is :
mp_physics = 3, 3, 3
ra_lw_physics = 1, 1, 1
ra_sw_physics = 1, 1, 1
radt = 9, 3, 1
sf_sfclay_physics = 2, 2, 2

Would this be the problem?
Thanks,

Chang
 
Chang,
I believe this issue is related to radiation. Can you try the option: radt = 9, 9, 9 and see whether the performance is better?
Please keep me updated. Thanks.
 
Top