This code must be compiled with OpenMP support as stated in README under onedrive submission.
Brief explanation:
Most execution of this file falls within 'nogtom' subroutine, involving multiple loops and microphysics. Only loops that are big enough that additional parallelizations we are adding would overcome their overhead are considered. The biggest loop is at line 753 (new file) which loops over all the divided grid and is responsible for multiple calculations. Directives are added after the first outer loop to maintain number accuracy. Some other loops that take considerable time were parallelized as well.
This code must be compiled with OpenMP support as stated in README under onedrive submission.
Brief explanation:
Most execution of this file falls within 'nogtom' subroutine, involving multiple loops and microphysics. Only loops that are big enough that additional parallelizations we are adding would overcome their overhead are considered. The biggest loop is at line 753 (new file) which loops over all the divided grid and is responsible for multiple calculations. Directives are added after the first outer loop to maintain number accuracy. Some other loops that take considerable time were parallelized as well.