Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Running the low-resolution CCSM3 / CESM 1.2 on a Linux workstation

I read in "The Low-Resolution CCSM3" by Yeager et al. (2006) that a Linux server with 16 processors was able to simulate 4-5 model years per day of clock time in CCSM3 using the T31x3 resolution. I conclude that it should be possible to run CCSM3 at low resolution on a Linux workstation today (using, say, a Xeon 26xx v3 CPU) and simulate a few model years per day, to do short, low-resolution runs and get to know the model. This is something I would like to try. I have two questions:1. Running CCSM3 at T31x3 resolution, what is a good ratio of processors for the atmosphere model and ocean model in order to run the model at near-optimal speed? Do the atmosphere and ocean require roughly equal numbers of processors at this resolution, or does one component model need siginificantly more computing power than the other?2. How much more computationally expensive is CESM 1.2, roughly speaking, when compared to CCSM3? Would it be possible to run CESM 1.2 on a modern workstation at T31_gx3v7 resolution and simulate 4-5 model years per day, as was possible with CCSM3 on a Linux server in 2006? Or is CESM so much more complex that this would be out of the question?Thank you for the information!
 
Top