Формалізований метод проектування застосувань в технології GPGPU
The method of researching systems with high-performance computing support, based on the transition systems apparatus (discrete computational model), is proposed. Two variants of synchronous product limitations of transition systems that model tha Nvidia CUDA approach are proposed. transition systems...
Збережено в:
| Дата: | 2018 |
|---|---|
| Автори: | , , |
| Формат: | Стаття |
| Мова: | Ukrainian |
| Опубліковано: |
Інститут програмних систем НАН України
2018
|
| Теми: | |
| Онлайн доступ: | https://pp.isofts.kiev.ua/index.php/ojs1/article/view/258 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Problems in programming |
| Завантажити файл: |  |
Репозиторії
Problems in programming| Резюме: | The method of researching systems with high-performance computing support, based on the transition systems apparatus (discrete computational model), is proposed. Two variants of synchronous product limitations of transition systems that model tha Nvidia CUDA approach are proposed. transition systems that represent two types of instructions, process of the warp instruction execution, and the process of warp scheduling were described. GPGPU application execution model was formalized and its correctness was proved. Two variants of the relevant Petri net which allowed automatic or semi-automatic detection of design errors were obtained.Problems in programming 2018; 2-3: 012-020 |
|---|