Benchmark of ADT-based solvers against procedural ones

Exactly what the title says. I think it will be important to test and benchmark the performance of FOODiE's solvers against their equivalents written in a procedural style, essentially testing the efficiency of abstract data types and overloaded operators. The results will have to be satisfactory in order for FOODiE to be considered in HPC applications. I am not sure yet what satisfactory means in this case but I will think about it.

Comparing the performance of FOODiE between major Fortran compilers will be elucidating as well.

Please discuss. I will be happy to make some tests in the next few weeks.