Skip to content

Latest commit

 

History

History
16 lines (12 loc) · 1.12 KB

README.md

File metadata and controls

16 lines (12 loc) · 1.12 KB
Raw

Linear-equations-solving-using-parallelism

Implentation of Back Substitution, Conjugate Gradient and Gauss Seidel using OpenMP parallelization

Solving large systems of linear equations in serial is time consuming and slow process.
Prallelizing the algorithms using suitable parallel constructs can provide the correct solutions with a lesser time complexity.

This project presents parallel and serial implementation of three algorithms for solving system of linear equations:

  1. Back Substitution,
  2. Conjugate Gradient and
  3. Gauss Seidel

In this project, OpenMP is used for parallelizing the algorithms.
For all the 3 algos, serial and parallel implementations is provided. Execution time for each of the 6 functions(serial and parallel implementations of the three algos) is calculated to determine and henceforth compare the efficiency of the three algorithms.

Two code files: required.c and random.c are basically the same implementation, but random.c considered a random matrix input, while required.c considers specific input matrix considering the limitations of the algorithms.