Paralelno programiranje s OpenMP bibliotekom

master's thesis

• Title (english): Parallel programming with the OpenMP library
• Author: Ivan Živković
• Mentor: Mladen Jurak (mentor)
• Committee member: Mladen Jurak (predsjednik povjerenstva)
• Committee member: Matej Mihelčić (član povjerenstva)
• Committee member: Tomislav Pejković (član povjerenstva)
• Committee member: Siniša Slijepčević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2024-03-01, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

U ovom radu dajemo pregled paralelnog programiranja s dijeljenom memorijom, kao i primjene OpenMP biblioteke na takvo programiranje. U prvom poglavlju obrađujemo programiranje s dretvama, paralelan pristup dijeljenim podatcima, sinkronizaciju dretvi te memorijske modele sustava s dijeljenom memorijom. Također se dotičemo programskog okruženja koje ćemo kasnije koristiti (programski jezik C++). U drugom poglavlju uvodimo OpenMP standard, dotičemo se njegovih ciljeva te dajemo kratki pregled standarda. Objašnjavamo kako OpenMP odgovara na potrebe paralelnog programiranja s dijeljenom memorijom. U trećem poglavlju dajemo primjere paralelizacije algoritama koristeći OpenMP. Kao primjer regularnog problema implementiramo metodu konjugiranih gradijenata za numeričko rješavanje linearnih sustava, a kao primjer neregularnog problema implementiramo algoritam za sortiranje quicksort. Testiramo skalabilnost obje implementacije izvršavanjem na paralelnom računalu s različitim brojevima dretvi i mjerenjem njihovih vremena izvršavanja

Abstract (english)

In this paper we provide an overview of shared memory parallel programming and the application of the OpenMP library to such programming. In the first chapter we go through multithreading, shared memory access, thread synchronization and shared-memory system memory models. We also touch upon the programming environment used later (the C++ programming language). In the second chapter we go through the basics of the OpenMP standard, touch upon its goals and give a short overview of the included functionality. We cover the use of OpenMP on shared memory parallel programming. In the third chapter we give examples of algorithm parallelization using OpenMP. As an example of a regular problem we implement the conjugate gradient method for numerical solving of systems of linear equations, and as an example of an irregular problem we implement the quicksort sorting algorithm. We test the scalability of both implementations by running them on a parallel computer with different thread counts and measuring their execution time

• Keywords: programiranje s dretvama; paralelan pristup dijeljenim podatcima; sinkronizaciju dretvi; algoritam za sortiranje quicksort
• Language: croatian
• URN:NBN: urn:nbn:hr:217:874294
• Study programme: Title: Computer Science and Mathematics; Study programme type: university; Study level: graduate; Academic / professional title: sveučilišni magistar računarstva i matematike (sveučilišni magistar računarstva i matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-03-14 12:50:03