Implementing the gauss seidel algorithm for solving eigenvalues of symmetric matrices with CUDA

Modern GPUs are more efficient than CPUs due to their highly parallel structure. The Gauss-Seidel algorithm is a method for solving the n linear equations of the form Ax=b, which uses previously computed results as soon as they are available. The Gauss Seidel algorithm is the modified method of Jacobi algorithm. The Gauss Seidel algorithm is used to solve the eigenvalues of the large matrices. In this project I will be creating API functions for Gauss Seidel method using cuda. In this project the Gauss Seidel Algorithm is implemented with CUDA on GPU to solve the eigenvalues of symmetric matrices. Initially Gauss Seidel algorithm is implemented on CPU and then is implemented on GPU and then their performances are checked. As GPU is a highly parallel structure with thousands of cores, the performance of GPU is faster than the CPU.