Fast Conjugade Gradient Solver

Implementation of Conjugate Gradient (CG) algorithm for solving sparse linear systems using MPI and CUDA. On a high level, the goal is to make full use of the available hardware resources and parallelise the CG algorithm using MPI (to make use multi-core CPUs) or CUDA (to make use available CUDA cores). See instructions for more detail about the problem. Each folder includes report about the implementation, and scaling properties of the implementation.