https://github.com/smillerc/coarray_field

A simple high-level field object in Fortran that manages domain decomposition via coarrays
https://github.com/smillerc/coarray_field

coarray-fortran fortran fortran2018

Last synced: about 1 month ago
JSON representation

A simple high-level field object in Fortran that manages domain decomposition via coarrays

• Host: GitHub
• URL: https://github.com/smillerc/coarray_field
• Owner: smillerc
• Created: 2020-10-26T19:10:54.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2020-12-03T15:01:56.000Z (about 4 years ago)
• Last Synced: 2024-01-30T09:11:49.579Z (12 months ago)
• Topics: coarray-fortran, fortran, fortran2018
• Language: Fortran
• Homepage:
• Size: 84 KB
• Stars: 3
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# A high level coarray field object to manage domain decomposition

* The goal of this project is to demonstrate how coarrays can be used to decompose a domain for physics-type simulations.

* This project starts off with the `Field` class from the book [Modern Fortran](https://github.com/modern-fortran/tsunami) by Milan Curcic

* Rudimentary field serialization is via the [`h5fortran`](https://github.com/geospace-code/h5fortran) library by Michael Hirsch

* Eventual OpenCL integration via the [`Focal`](https://github.com/LKedward/focal) library by Laurence Kedward 

## Build instructions

```bash

mkdir build

cd build

cmake .. -DCMAKE_BUILD_TYPE="Debug" \

    -DENABLE_COARRAY=ON \

    -DUSE_OPENMP_THREADS=NO \

    -DUSE_OPENMP_SIMD=YES \

    -DUSE_OPENCL=NO \

    -DENABLE_TESTING=YES

```

## Example Use [WIP]

```fortran

type(field_2d_t) :: rho, u, rho_u

rho = field_2d(name='rho', long_name='density', &

               descrip='Cell-centered mass density', &

               units='g/cc', global_dims=[20, 20], n_halo_cells=2)

u = field_2d(name='u', long_name='x-velocity', &

               descrip='Cell-centered x-velocity', &

               units='cm/s', global_dims=[20, 20], n_halo_cells=2)

! Create a new field based off of existing ones

rho_u = rho * u

! Get max

call rho_u%maxval()

```