Vzporedni programski jeziki namesto vzporednih programskih ogrodij

Novosel, Rok (author), Slivnik, Boštjan (mentor)

Čeprav vzporedni programski jeziki obstajajo že desetletja, v znanstvenem vzporednem programiranju še vedno prevladujejo Fortran, C in C++ dopolnjeni z vzporednimi programskimi ogrodji MPI, OpenMP, OpenCL in CUDA. V tem delu izvedemo primerjalno študijo novih vzporednih programskih jezikov Chapel in Julia. Glede na vzporedne zmožnosti se oba jezika precej razlikujeta med sabo in v primerjavi s Fortranom in Cjem. Študijo izvedemo na testnih problemih, ki izpostavljajo potrebo po različnih pristopih vzporednega programiranja. Testne probleme implementiramo v Chapelu in Juliji ter Cju, ki ga dopolnimo z MPI-jem in OpenMP-jem. Pokažemo, da tako Chapel kot Julia predstavljata uspešni alternativi Fortranu in C/C++ z vzporednimi programskimi ogrodji. Učinkovitost programerja se znatno izboljša, medtem ko hitrost programov ni bistveno poslabšana.

• Language: Slovenian
• Keywords: programski jeziki, vzporedno programiranje, vzporedni programski jeziki
• Work type: Master's thesis/paper
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2019
• PID: 20.500.12556/RUL-109336
• COBISS.SI-ID: 1538306499
• Publication date in RUL: 30.08.2019

Secondary language

• Language: English
• Title: Parallel programming languages instead of parallel programming frameworks

Abstract

Although parallel programming languages have existed for decades, (scientific) parallel programming is still dominated by Fortran and C/C++ augmented with parallel programming frameworks, e.g., MPI, OpenMP, OpenCL and CUDA. We perform a comparative study of Chapel and Julia, two languages quite different from one another as well as from Fortran and C, in regard to parallel programming on distributed and shared memory computers. The study is carried out using test cases that expose the need for different approaches to parallel programming. Test cases are implemented in Chapel and Julia, and in C augmented with MPI and OpenMP. It is shown that both languages, Chapel and Julia, represent a viable alternative to Fortran and C/C++ augmented with parallel programming frameworks: the programmer's efficiency is considerably improved while the speed of programs is not significantly affected.

• Keywords: programming languages, parallel programming, parallel programming languages