midhun_ndr@gmail_com
New Member
Hi, I am running CESM 1.2 with B1850C5 compset. I have 12 nodes available with 20 cores per node (http://www.nlhpc.cl/es/cluster2014/) making a total of 240 cores.Currently, I make simulations by giving 240 cores to each components. Here is my PE layout.
}
Overall I get a modelthroughput of 6.4myears/wday. Here is the component wise breakup.
TOT Run Time: 13483.118 seconds 36.940 seconds/mday 6.41 myears/wday
LND Run Time: 78.358 seconds 0.215 seconds/mday 1102.63 myears/wday
ROF Run Time: 7.841 seconds 0.021 seconds/mday 11019.00 myears/wday
ICE Run Time: 1838.970 seconds 5.038 seconds/mday 46.98 myears/wday
ATM Run Time: 6411.420 seconds 17.566 seconds/mday 13.48 myears/wday
OCN Run Time: 3186.190 seconds 8.729 seconds/mday 27.12 myears/wday
GLC Run Time: 0.000 seconds 0.000 seconds/mday 0.00 myears/wday
WAV Run Time: 0.000 seconds 0.000 seconds/mday 0.00 myears/wday
CPL Run Time: 963.904 seconds 2.641 seconds/mday 89.64 myears/wdayHow can I optimize the model throughput?
CAM5 is the most expensive component in the compset, but total model throughput is about only half of CAM5 throughput. Why is it so?
}
Overall I get a modelthroughput of 6.4myears/wday. Here is the component wise breakup.
TOT Run Time: 13483.118 seconds 36.940 seconds/mday 6.41 myears/wday
LND Run Time: 78.358 seconds 0.215 seconds/mday 1102.63 myears/wday
ROF Run Time: 7.841 seconds 0.021 seconds/mday 11019.00 myears/wday
ICE Run Time: 1838.970 seconds 5.038 seconds/mday 46.98 myears/wday
ATM Run Time: 6411.420 seconds 17.566 seconds/mday 13.48 myears/wday
OCN Run Time: 3186.190 seconds 8.729 seconds/mday 27.12 myears/wday
GLC Run Time: 0.000 seconds 0.000 seconds/mday 0.00 myears/wday
WAV Run Time: 0.000 seconds 0.000 seconds/mday 0.00 myears/wday
CPL Run Time: 963.904 seconds 2.641 seconds/mday 89.64 myears/wdayHow can I optimize the model throughput?
CAM5 is the most expensive component in the compset, but total model throughput is about only half of CAM5 throughput. Why is it so?