The Survey of Methodologies, Approaches, and Challenges in Parallel Programming Using High-Performance Computing Systems paper, published in 2020, explores some of the state-of-the-art methodologies and solutions supporting the development of parallel applications for modern high-performance computing systems.
|Survey of Methodologies, Approaches, and Challenges in Parallel Programming Using High-Performance Computing Systems
|Pedro Valero-Lara, Paweł Czarnul, Jerzy Proficz and Krzysztof Drypczewski
|2020, Scientific Programming Journal, DOI: 10.1155/2020/4176794
|~8,500 words, 45 min. read
The Survey of Methodologies, Approaches, and Challenges in Parallel Programming Using High-Performance Computing Systems paper was published in 2020 in the Scientific Programming Open Access Journal. This article provides a good start for identifying and understanding some current trends in high-performance computing and some of the challenges to be addressed in the near future.
The paper provides a review of some of the contemporary methodologies and APIs for parallel programming. It also gives examples of representational technologies and classifies them by target system type, memory types, one-sided/two-sided communication patterns, and programming abstraction level. The representative technologies are further analyzed based on other criteria such as programming model, languages, supported platforms, license, optimization goals, ease of programming, debugging, deployment, portability, level of parallelism etc.
BSSw.io readers, who have some scientific computing background but are relatively new to HPC and are short on time, can glean a high-level overview of HPC methodologies and practices from this paper. The reference list also provides some resources for delving deeper into each of the specific topics mentioned in the paper.