Skip to content

Latest commit

 

History

History
29 lines (18 loc) · 1.26 KB

README.md

File metadata and controls

29 lines (18 loc) · 1.26 KB

Parallel Programming Project 2024

This project focuses on parallelizing a sequential multi-start multidimensional scaling (MDS) algorithm (multistart_mds_seq.c) using different parallel programming paradigms: OpenMP, OpenMP tasks, and MPI. The aim is to explore and compare the performance and scalability of these parallelization techniques when applied to a computationally intensive algorithm.

Files and Parallelization Techniques

  1. multistart_mds_omp.c:

    • Parallelization Method: OpenMP

  2. multistart_mds_omp_tasks.c:

    • Parallelization Method: OpenMP Tasks

  3. multistart_mds_mpi.c:

    • Parallelization Method: MPI (Message Passing Interface)

Key Objectives

  • Performance Analysis: Measure and compare the performance of each parallelization technique, focusing on speedup, efficiency, and scalability.
  • Scalability Testing: Evaluate how each method scales with increasing numbers of processors/threads.
  • Optimization: Investigate potential optimizations in each parallelization method to maximize performance.

How to Run

Run make on the directory.

Dependencies

  • OpenMP: Requires a compiler with OpenMP support (e.g., GCC, Clang).
  • MPI: Requires an MPI implementation (e.g., MPICH, OpenMPI).