Scribd
Upload
17 views2 pages

Task PDF

The document outlines a task for SIT315 Programming Paradigms focused on parallel matrix multiplication using C/C++. It requires the implementation of both sequential and parallel matrix multiplication programs, evaluation of their performance, and a comparison of results using OpenMP and pthreads. Submission includes source code, documentation of parallelization, and performance evaluations across different input sizes and thread counts.

Uploaded by

rhythmspotify123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
17 views2 pages

Task PDF

The document outlines a task for SIT315 Programming Paradigms focused on parallel matrix multiplication using C/C++. It requires the implementation of both sequential and parallel matrix multiplication programs, evaluation of their performance, and a comparison of results using OpenMP and pthreads. Submission includes source code, documentation of parallelization, and performance evaluations across different input sizes and thread counts.

Uploaded by

rhythmspotify123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

M2.T1P_Parallel_Matrix_Multiplication.

md

SIT315 Programming Paradigms


Module2 Concurrent Programming
TaskM2.T1P: Parallel Matrix Multiplication
Overview of the task
To fulfill the requirements of this task, you will need to demonstrate your skills to use multi-threading in
C/C++ to speed up sequential program. In this task, we use matrix multiplication as the example problem.
See here (https://en.wikipedia.org/wiki/Matrix_multiplication)
Submission Details
Please make sure to provide the following:
Source code of the sequential matrix multiplication program,
Document outlining your program parallelisation and decomposition,
Source code of the parallel program,
Evaluation of your program on different input sizes and number of threads,
Source code of the OpenMP version, and
Updated evaluation document comparing OpenMP vs the other two programs.
Instructions
. Implement a simple matrix multiplication program in C or C++. Matrix multiplication of C = A X B
where A, B and C are matrices of size N X N (N rows and N columns) and both A and B are
initialised with random values. At the end of the program, You should writ your output to a file. How
matrix multiplication works? Please see here (https://en.wikipedia.org/wiki/Matrix_multiplication).
Simply to calculate Ci,j (row i, column j), you multiply every element in matrix A row i by every
element in matrix B column j. This is simply a three nested loops.
. At the end of the program, please print the execution time - time taken to calculate the matrix
multiplication - not including initialisation of matrices or writing results to file.
. Once you have completed and tested the program, please review your code and develop a
roadmap to parallelise your code. You should start with decomposition of the program/problem into
sub-tasks - i.e. partioning data/tasks. Document your list of sub-tasks or activities you plan to do
in parallel vs activities that need to be in sequence.
. Implement your parallel algorithm in C or C++ using pthread library.
. Evaluate the performance of your program (using execution time as a metric), to assess the speed
up achieved. Reflect on different sizes of the input matrices and also number of threads you used
in your program - vary from 2 to MAX number of threads. Compare the results with the sequential
program.
. Modify your sequential program to use OpenMP to achieve parallelism
1/2
M2.T1P_Parallel_Matrix_Multiplication.md

. Evaluate the performance of the OpenMP implementation vs pthread implementation vs the


sequential program. Discuss your findings.
. Submit your task as detailed on the submission details section above to OnTrack.

2/2

You might also like