Date Approved

12-12-2024

Graduate Degree Type

Project

Degree Name

Applied Computer Science (M.S.)

Degree Program

School of Computing and Information Systems

First Advisor

Christian Trefftz

Academic Year

2024/2025

Abstract

This project investigates the implementation and optimization of parallel computing techniques using GPU acceleration frameworks such PyCuda. With the increasing demand for high-performance solutions in data-intensive applications, GPUs offer a compelling alternative to traditional CPU-based processing. The primary objective of this work is to harness the computational power of GPUs to achieve significant performance enhancements for complex workloads. The study focuses on two essential research questions: How effectively can PyCuda accelerate computational tasks, and what measurable performance gains can be achieved compared to CPU-based implementations? By addressing these questions, this project explores the applicability of GPU programming in tasks such as matrix operations, image processing, and sorting algorithms. Through systematic implementation and optimization, the results demonstrate that GPU-accelerated approaches yield substantial speedups, with improvements ranging from 10x to 20x for matrix computations and noticeable reductions in execution time for image processing tasks like convolutions and edge detection. Additionally, the project evaluates the ease of use and efficiency PyCuda in their respective programming ecosystems. PyCuda simplifies GPU programming with Python’s dynamic and versatile framework. The study further highlights techniques such as shared memory utilization and kernel optimization that enhance GPU efficiency. Overall, this project contributes to the understanding of GPU acceleration frameworks, providing a foundation for further exploration in high-performance computing. The findings have implications for fields requiring rapid computation, such as machine learning, big data processing, and scientific simulations, making GPUs an indispensable tool in modern computational research.

Download

Share

COinS