论文信息 - Development and performance of a new version of the OASIS coupler , OASIS 3-MCT _ 3 . 0

Development and performance of a new version of the OASIS coupler , OASIS 3-MCT _ 3 . 0

OASIS is coupling software developed primarily for use in the climate community. It provides 10 the ability to couple different models with low implementation and performance overhead. OASIS3-MCT is the latest version of OASIS. It includes several improvements compared to OASIS3 including elimination of a separate hub coupler process, parallelization of the coupling communication and run time grid interpolation, and the ability to easily reuse mapping weights files. OASIS3-MCT_3.0 is the latest release and includes the ability to couple between components running sequentially on the same set of tasks 15 as well as to couple within a single component between different grids or decompositions such as physics, dynamics, and I/O. OASIS3-MCT has been tested with different configurations on up to 32,000 processes, with components running on high-resolution grids with up to 1.5 million grid cells, and with over 10,000 two dimensional coupling fields. Several new features will be available in OASIS3-MCT_4.0 and some of those are also described. 20

[1] Olivier Thual,et al. Climatology and interannual variability simulated by the ARPEGE-OPA coupled model , 1995 .

[2] S. Valcke,et al. The OASIS3 coupler: a European climate modelling community software , 2012 .

[3] Mariana Vertenstein,et al. A new flexible coupler for earth system modeling developed for CCSM4 and CESM1 , 2012, Int. J. High Perform. Comput. Appl..

[4] Arthur A. Mirin,et al. Improving the performance scalability of the community atmosphere model , 2012, Int. J. High Perform. Comput. Appl..

[5] J. Larson,et al. M × N Communication and Parallel Interpolation in CCSM 3 Using the Model Coupling Toolkit , 2005 .

[6] Jay Walter Larson,et al. The Model Coupling Toolkit: A New Fortran90 Toolkit for Building Multiphysics Parallel Coupled Models , 2005, Int. J. High Perform. Comput. Appl..

[7] Graham D. Riley,et al. Sharing experiences and outlook on Coupling Technologies for Earth System Models , 2016 .

[8] Jesper Larsson Träff,et al. MPI on a Million Processors , 2009, PVM/MPI.