Choosing a library for the Python programming language for visualizing the operation of parallel algorithms

• Published in: E3S web of conferences Vol. 508; p. 3002
• Main Authors: Sharov, Sergii, Sitsylitsyn, Yurii, Naumuk, Oleksii, Lubko, Dmytro, Kolmakova, Vira
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Les Ulis EDP Sciences 01.01.2024
• Subjects: Algorithms; Animation; C plus plus; Libraries; Machine learning; Object oriented programming; parallel computing; Parallel programming; Pedagogy; programming; Programming languages; Python; Real time; Software; Ukraine
• ISSN: 2267-1242; 2555-0403; 2267-1242
• DOI: 10.1051/e3sconf/202450803002

The research compares the capabilities of several libraries for the Python language, which allow creating a test application and visually demonstrate the operation of a parallel program in real time. It was found that the Python language is often used to develop parallel programs with internal and external libraries. To provide multithreading and parallelism, applications created in Python use external libraries, including mpi4py.futures, PETSc for Python, MPI for Python, d2o, Playdoh, PyOMP, and others. Visualization and animation of the operation of parallel programs will help to understand the principles of parallel computing. We compared test applications created with the use of Matplotlib, Seaborn, Plotly, Bokeh, Pygame, PyOpenGL libraries. According to the results of the observation, it was found that the Seaborn library is the best choice for developing a test application for animating the operation of a parallel program.