Proposal for Semi-Automatic Parallelism of Linear Pipelines in C++
Abstract
Atingir alto desempenho, portabilidade e produtividade continua sendo um desafio fundamental na comunidade de computação de alto desempenho. Entretanto, a paralelização automática de programas nunca foi amplamente adotada devido à sua inconsistência. Assim, a tarefa onerosa da programação paralela permanece sob a responsabilidade direta dos programadores. Esse trabalho propõe-se estudar técnicas de paralelismo com compiladores.
References
Bhosale, A. and et al. (2022). Automatic and interactive program parallelization using the cetus source to source compiler infrastructure v2.0. Electronics, 11:1–22.
Harel, R. and et al. (2019). Source-to-source parallelization compilers for scientific shared-memory multi-core and accelerated multiprocessing: Analysis, pitfalls, enhancement and potential. International Journal of Parallel Programming, 48:1–31.
Tournavitis, G. and Franke, B. (2010). Semi-automatic extraction and exploitation of hierarchical pipeline parallelism using profiling information. In PACT, page 377–388, Vienna, Austria. Association for Computing Machinery.
Harel, R. and et al. (2019). Source-to-source parallelization compilers for scientific shared-memory multi-core and accelerated multiprocessing: Analysis, pitfalls, enhancement and potential. International Journal of Parallel Programming, 48:1–31.
Tournavitis, G. and Franke, B. (2010). Semi-automatic extraction and exploitation of hierarchical pipeline parallelism using profiling information. In PACT, page 377–388, Vienna, Austria. Association for Computing Machinery.
Published
2025-04-23
How to Cite
HOFFMANN, Renato B.; GRIEBLER, Dalvan.
Proposal for Semi-Automatic Parallelism of Linear Pipelines in C++. In: REGIONAL SCHOOL OF HIGH PERFORMANCE COMPUTING FROM SOUTHERN BRAZIL (ERAD-RS), 25. , 2025, Foz do Iguaçu/PR.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 149-150.
ISSN 2595-4164.
DOI: https://doi.org/10.5753/eradrs.2025.6803.
