NAS Parallel Benchmarks with Python: a performance and programming effort analysis focusing on GPUs

• Published in: The Journal of supercomputing Vol. 79; no. 8; pp. 8890 - 8911
• Main Authors: Di Domenico, Daniel, Lima, João V. F., Cavalheiro, Gerson G. H.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.05.2023; Springer Nature B.V
• Subjects: Algorithms; Allocations; Artificial intelligence; Benchmarks; C plus plus; C++ (programming language); Compilers; Computer Science; Experiments; Fourier transforms; High level languages; Interpreters; Machine learning; Monte Carlo simulation; Partial differential equations; Processor Architectures; Programming; Programming Languages; Python; NPB; Numba; GPU; Programming effort; Python
• ISSN: 0920-8542; 1573-0484; 1573-0484
• DOI: 10.1007/s11227-022-04932-3

Compiled low-level languages, such as C/C++ and Fortran, have been employed as programming tools to implement applications to explore GPU devices. As a counterpoint to that trend, this paper presents a performance and programming effort analysis with Python, an interpreted and high-level language, which was applied to develop the kernels and applications of NAS Parallel Benchmarks targeting GPUs. We used Numba environment to enable CUDA support in Python, a tool that allows us to implement the GPU programs with pure Python code. Our experimental results showed that Python applications reached a performance similar to C++ programs employing CUDA and better than C++ using OpenACC for most NPB benchmarks. Furthermore, Python codes demanded less operations related to the GPU framework than CUDA, mainly because Python needs a lower number of statements to manage memory allocations and data transfers. Despite that, our Python implementations required more operations than OpenACC ones.