Run parallel NUCOM

Share:

Description

Code to run NUCOMBog parallel on multiple cores.

Usage

1
2
runparallelNUCOM(setup, clustertype, numCores = 1, parameters,
  MPI_server = F)

Arguments

setup

The setup needs to be made before by running the setup_NUCOM function.

clustertype

Clustertype: The model has only been tested on SOCK cluster, which is the set to default.

numCores

Number of Cores on which are model needs to be run (NOTE: Non-parallel runs can only be run on 1 core). Default is 1.

parameters

The parameters which are used in the model. If no parameter values are given the default values will be used. The parameters have to have the format of a dataframe with colum names: "names" and "values", see example https://github.com/jeroenpullens/NUCOMBog_data. The default parameters are from Heijmans et al. 2008.

MPI_server

set to TRUE if after each iteration the MPI server needs to be destroyed. Default is FALSE

Author(s)

JWM Pullens

Source

The test data can be downloaded from http://jeroenpullens.github.io/NUCOMBog_data/. And the executable of the model can downloaded from https://github.com/jeroenpullens/source_modelMEE.

References

Heijmans, M., Mauquoy, D., van Geel, B., and Berendse, F. (2008). Long-term effects of climate change on vegetation and carbon dynamics in peat bogs. Journal of Vegetation Science, 19(3)

Examples

1
2
3
4
5
6
7
8
9
## Not run: 
!!the variable "test_setup" is from the function setupNUCOM, see the help for more information!!

parallel<-runparallelNUCOM(setup = test_setup,
                            clustertype = "SOCK",
                            numCores = 1,
                            parameters=initialParameters)

## End(Not run)