bsuirДоклады БГУИРDoklady BGUIR1729-76482708-0382БГУИРbsuir-829Research ArticleСтатьиЭВРИСТИЧЕСКИЙ ГЕНЕТИЧЕСКИЙ АЛГОРИТМ ОПТИМИЗАЦИИ ВЫЧИСЛИТЕЛЬНЫХ КОНВЕЙЕРОВHeuristic genetic algorithm of computational pipelines optimizationПрихожийА. А.PrihozhyA. A.noemail@neicon.ruЖдановскийА. М.ZhdanouskiA. M.noemail@neicon.ruКарасикО. Н.KarasikO. N.noemail@neicon.ruМаттавеллиМ. ..MattavelliM. ..noemail@neicon.ruБелорусский национальный технический университетBelarusian National Technical UniversityÉcole polytechnique fédérale de LausanneÉcole polytechnique fédérale de Lausanne201703062019013441Copyright © Прихожий А.А., Ждановский А.М., Карасик О.Н., Маттавелли М..., 20192019Прихожий А.А., Ждановский А.М., Карасик О.Н., Маттавелли М...Prihozhy A.A., Zhdanouski A.M., Karasik O.N., Mattavelli M...Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://doklady.bsuir.by/jour/article/view/829

Проведен анализ вычислительных конвейеров и методов их оптимизации. Рассмотрен класс потоковых конвейеров, не использующих разделение вычислительных ресурсов и достигающих наибольшей производительности. Для этого класса рассмотрены задачи оптимизации конвейеров, решаемые для случайной логики в процессе высокоуровневого синтеза по поведенческим спецификациям на языках программирования и описания аппаратуры. Исследован эвристический генетический алгоритм оптимизации, применимый к проектам большого размера.

An analysis of computational pipelines and their optimization methods has been performed. A class of dataflow pipelines that do not use resource sharing and obtain high throughput has been considered. Pipeline optimization tasks being solved during high-level synthesis from random logic behavioral specifications in programming and hardware description languages have been considered. A heuristic genetic optimization algorithm which is capable of handling large designs has been proposed and investigated.

поток данныхвычислительный конвейервысокоуровневый синтезоптимизацияdata flowcomputational pipelinehigh-level synthesisoptimizationReferencesScheduling and hardware sharing in pipelined data paths / K.S. Hwang [et. al.] // Proc. ICCAD-89, November 1989. P. 24-27.Scheduling and hardware sharing in pipelined data paths / K.S. Hwang [et. al.] // Proc. ICCAD-89, November 1989. P. 24-27.Sun W., Wirthlin M., Neuendorffer S. FPGA pipeline synthesis design exploration using module selection and resource sharing // Trans. Comp.-Aided Des. Integ. Cir. Sys. 2007. Vol. 26, No. 2. P. 254-265.Sun W., Wirthlin M., Neuendorffer S. FPGA pipeline synthesis design exploration using module selection and resource sharing // Trans. Comp.-Aided Des. Integ. Cir. Sys. 2007. Vol. 26, No. 2. P. 254-265.Bakshi S., Gajski D. Component Selection for High-Performance Pipelines // IEEE Trans. VLSI Syst. 1996. Vol. 4, No. 2. P. 181-194.Bakshi S., Gajski D. Component Selection for High-Performance Pipelines // IEEE Trans. VLSI Syst. 1996. Vol. 4, No. 2. P. 181-194.Prihozhy A. High-level Synthesis through Transforming VHDL Models // System-on-Chip Methodologies and Design Languages. 2001. P. 135-146.Prihozhy A. High-level Synthesis through Transforming VHDL Models // System-on-Chip Methodologies and Design Languages. 2001. P. 135-146.Dataflow/actor-oriented language for the design of complex signal processing systems / C. Lucarz [et. al.] // Proc. Conf. on Design and Architectures for Signal and Image processing, November 2008. P. 1-8.Dataflow/actor-oriented language for the design of complex signal processing systems / C. Lucarz [et. al.] // Proc. Conf. on Design and Architectures for Signal and Image processing, November 2008. P. 1-8.Synthesis and Optimization of Pipelines for HW Implementations of Dataflow Programs / A. Prihozhy [et. al.] // IEEE Trans. on CAD of Integrated Circuits and Systems. 2016. Vol. 34, No. 10. P. 1613-1626.Synthesis and Optimization of Pipelines for HW Implementations of Dataflow Programs / A. Prihozhy [et. al.] // IEEE Trans. on CAD of Integrated Circuits and Systems. 2016. Vol. 34, No. 10. P. 1613-1626.Rahman H., Prihozhy A., Mattavelli M. Pipeline Synthesis and Optimization of FPGA-Based Video Processing Applications with CAL // EURASIP Journal on Image and Video Processing. 2011. Vol. 19. P. 1-28.Rahman H., Prihozhy A., Mattavelli M. Pipeline Synthesis and Optimization of FPGA-Based Video Processing Applications with CAL // EURASIP Journal on Image and Video Processing. 2011. Vol. 19. P. 1-28.

The authors declare that there are no conflicts of interest present.