[RegCNET] RegCM 4 - HadGEM2 RCP4.5 data taking more run time

[egy_mohamad@yahoo.com]

Dear Raja,
The problem is not in CFL criteria or your model will crash before completing your run. The time you mentioned in your email is reasonable because the resolution. When you run the with very high resolution and with a big number of grids this will take more time. Because the idea is not the resolution but the number of the calculating operations done by the model. Hope this is helpfull.

Sent from my HTC

----- Reply message -----
From: "Rajalakshmi Dhandapani" <rajiamt at yahoo.in>
To: "Mujtaba" <mhassan512 at yahoo.com>, "'regcnetusers'" <regcnet at lists.ictp.it>
Subject: [RegCNET] RegCM 4 - HadGEM2 RCP4.5 data taking more run time
Date: Thu, Feb 20, 2014 9:11 AM
Dear Mujtaba,
Thanks for your suggestions. But in the run what we made for ds=60, dt=100 was used. Though dt is not three times greater than ds, time taken is more.
regcm.in file used in simulation of 60km ds is attached. Kindly find in the attachment and give suggestions.
--
Thanks and regards,
Dr. D. Rajalakshmi,
Research Associate,
Agro Climate Research Centre,
Tamil Nadu Agricultural University,
Coimbatore-641003.



On Thursday, 20 February 2014 12:16 PM, Mujtaba
<mhassan512 at yahoo.com> wrote:
Dear Dr.Rajalakshmi,This depend upon the value of dt that you have set in timeparam stanza of input namelist of the RegCM.A good rule of thumb is to have a dt not greater than three times the ds value in km specified in the geoparam stanza. A greater value may lower the computational time, but in case of strong advection may lead to non-accurate computation or even the violation of CFL condition and the divergence of the solution.  Best Regards    From: regcnet-bounces at lists.ictp.it [mailto:regcnet-bounces at lists.ictp.it] On Behalf Of Rajalakshmi DhandapaniSent: Thursday, February 20, 2014 2:01 PMTo: regcnetusersSubject: [RegCNET] RegCM 4 - HadGEM2 RCP4.5 data taking more run time  Dear all, I am using RegCM4.4 rc22 for model run over India using HadGEM2 RCP 4.5 dataset. Run was tried for 60km and 25 km. Model is taking 1 hour to complete one day for 25km run and 45 minutes for 60 km run. I have to run the model from 1971-2100. Can anyone suggest the reason why the model is taking so much time for a day. Is there a way to reduce the time taken by the model?Kindly give suggestions regarding the issue.Thanks and regards--Dr. D. Rajalakshmi,Research Associate,Agro Climate Research Centre,Tamil Nadu Agricultural University,Coimbatore - 641003.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ictp.it/pipermail/regcnet/attachments/20140220/bcbb2ac8/attachment.htm>