/****************************************************/

From Serial to Parallel: A simple training using the Martix-Vector multiplication algorithm

Directory layout(tree -d):

Training
├── External_Functions
├── GPUs
├── Multinode
│   ├── Hybrid
│   │   ├── Cuda-MPI
│   │   ├── OpenMP-MPI
│   │   └── PHI-MPI
│   └── MPI
├── OpenMP
├── Outputs
│   └── Debug
└── Serial

Loader contains required modules for all makefiles. Create executables with 'make' in corresponding directories. 

/****************************************************/