Abordagem e representação gráfica para visualização de hierarquia entre regiões e desequilíbrio de carga de processamento em programas paralelos
High-performance computing environments are constantly evolving, becoming increasingly important for solving problems in the most diverse areas of science and engineering. For developers to know if they are taking advantage of all the resources that these environments have to offer when using the...
Na minha lista:
| Autor principal: | |
|---|---|
| Outros Autores: | |
| Formato: | bachelorThesis |
| Idioma: | pt_BR |
| Publicado em: |
Universidade Federal do Rio Grande do Norte
|
| Assuntos: | |
| Endereço do item: | https://repositorio.ufrn.br/handle/123456789/48819 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: | High-performance computing environments are constantly evolving, becoming increasingly
important for solving problems in the most diverse areas of science and engineering. For
developers to know if they are taking advantage of all the resources that these environments
have to offer when using them to run their programs, it is important that there is a way
to analyze the performance obtained. Thus, performance analysis tools become essential
for developers to identify performance bottlenecks and optimize their code. This work
proposes the refinement of a visualization tool that allows the analysis of parallel scalability
trends of a program in a simple and intuitive way. The enhancement was made with the
addition of several important functionalities. The objective is to allow the possibility for
parallel computing developers to analyze the performance of their parallel programs more
quickly without worrying about organizing execution data manually. |
|---|