FAMOUS, faster: using parallel computing techniques to accelerate the FAMOUS/HadCM3 climate model with a focus on the radiative transfer algorithm

This work is distributed under the Creative Commons Attribution 3.0 License.-- et al.We have optimised the atmospheric radiation algorithm of the FAMOUS climate model on several hardware platforms. The optimisation involved translating the Fortran code to C and restructuring the algorithm around the computation of a single air column. Instead of the existing MPI-based domain decomposition, we used a task queue and a thread pool to schedule the computation of individual columns on the available processors. Finally, four air columns are packed together in a single data structure and computed simultaneously using Single Instruction Multiple Data operations.Sony CSL would like to thank the UK Met Office for providing us with a Vendor Benchmarking License, and the Sony Computer Entertainment R&D teams for their support. This work was also supported by EU FP6 project European Climate of the Last Millennium (MILLENNIUM 017008). O. B. and N. B. acknowledge support from the DECC/Defra Hadley Centre Climate Programme (GA01101).Peer reviewe