maisМоделирование и анализ информационных системModeling and Analysis of Information Systems1818-10152313-5417Yaroslavl State University10.18255/1818-1015-2021-2-198-214mais-1488Research ArticleTheory of ComputingТрансформация функционально-потоковых параллельных программ в императивныеThe System for Transforming the Code of Dataflow Programs into Imperativehttps://orcid.org/0000-0002-3340-6678ВасильевВладимир СергеевичVasilevVladimir S.vsvasilev@sfu-kras.ruhttps://orcid.org/0000-0002-5487-0699ЛегаловАлександр ИвановичLegalovAlexander I.alegalov@hse.ruhttps://orcid.org/0000-0002-2115-5461ЗыковСергей ВикторовичZykovSergey V.szykov@hse.ruСибирский федеральный университетРоссияSiberian Federal UniversityRussian FederationНациональный исследовательский университет «Высшая школа экономики»РоссияHigher school of Economics, National research UniversityRussian Federation202111062021282198214Copyright © Васильев В.С., Легалов А.И., Зыков С.В., 20212021Васильев В.С., Легалов А.И., Зыков С.В.Vasilev V.S., Legalov A.I., Zykov S.V.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://www.mais-journal.ru/jour/article/view/1488

Функционально-потоковая парадигма параллельного программирования ориентирована на разработку параллельных переносимых программ. Исходный код функционально-потоковых программ транслируется в набор графов, отражающих информационные и управляющие зависимости. Основным способом их исполнения является интерпретация, что не позволяет эффективно выполнять вычисления на реальных параллельных вычислительных системах и ведет к низкой производительности. Для непосредственного выполнения программ на существующих вычислительных системах требуется использование специфических методов оптимизации и трансформации, учитывающих особенности как языка программирования, так и архитектуры исполнителя. В настоящее время наиболее распространенной является архитектура Фон-Неймана, параллельное программирование для которой в большинстве случаев осуществляется с использованием языков, поддерживающих императивный стиль и ориентированных на статическую систему типов. Для различных архитектур параллельных вычислительных систем существуют разнообразные подходы к написанию параллельных программ. Трансформация функционально-потоковых параллельных программ в императивные позволяет сформировать общий каркас из фрагментов императивного кода, непосредственно отображающих последовательные вычисления, который в дальнейшем может быть адаптирован к конкретной параллельной архитектуре. В работе рассматривается подход к выполнению такого типа трансформации, заключающийся в выделении фрагментов функционально-потоковых параллельных программ в качестве шаблонов, заменяемых впоследствии на эквивалентные фрагменты императивных языков. Предлагаемые методы трансформации позволяют порождать программный код, к которому в дальнейшем можно применять различные оптимизирующие преобразования, включая распараллеливание с учетом целевой архитектуры.

Functional dataflow programming languages are designed to create parallel portable programs. The source code of such programs is translated into a set of graphs that reflect information and control dependencies. The main way of their execution is interpretation, which does not allow to perform calculations efficiently on real parallel computing systems and leads to poor performance. To run programs directly on existing computing systems, you need to use specific optimization and transformation methods that take into account the features of both the programming language and the architecture of the system. Currently, the most common is the Von Neumann architecture, however, parallel programming for it in most cases is carried out using imperative languages with a static type system. For different architectures of parallel computing systems, there are various approaches to writing parallel programs. The transformation of dataflow parallel programs into imperative programs allows to form a framework of imperative code fragments that directly display sequential calculations. In the future, this framework can be adapted to a specific parallel architecture. The paper considers an approach to performing this type of transformation, which consists in allocating fragments of dataflow parallel programs as templates, which are subsequently replaced by equivalent fragments of imperative languages. The proposed transformation methods allow generating program code, to which various optimizing transformations can be applied in the future, including parallelization taking into account the target architecture.

трансформация программфункционально-потоковое параллельное программированиеанализ программтипизацияпромежуточные представления программtransformation of programsdataflow parallel programmingprogram analysistypingintermediate program representationsReferencesK. Vivek, Parallel Computing Architectures and APIs: IoT Big Data Stream Processing. 2019.K. Vivek, Parallel Computing Architectures and APIs: IoT Big Data Stream Processing. 2019.I. I. Levin, A. I. Dordopulo, and V. A. Gudkov, “Programming of reconfigurable computing nodes in the COLAMO language,” Training manual. Taganrog: Publishing house of TTI SFEDU. In Russian, 2011.I. I. Levin, A. I. Dordopulo, and V. A. Gudkov, “Programming of reconfigurable computing nodes in the COLAMO language,” Training manual. Taganrog: Publishing house of TTI SFEDU. In Russian, 2011.A. I. Dordopulo and I. I. Levin, “Resource-independent programming of hybrid reconfigurable computing systems,” in Russian Supercomputing Days. In Russian, 2017, pp. 714-723.A. I. Dordopulo and I. I. Levin, “Resource-independent programming of hybrid reconfigurable computing systems,” in Russian Supercomputing Days. In Russian, 2017, pp. 714-723.V. Kasyanov, “Sisal 3.2: functional language for scientific parallel programming,” Enterprise Information Systems, vol. 7, no. 2, pp. 227-236, 2013.V. Kasyanov, “Sisal 3.2: functional language for scientific parallel programming,” Enterprise Information Systems, vol. 7, no. 2, pp. 227-236, 2013.A. I. Legalov, “Functional language for creating architecturally independent parallel programs,” Computing technologies, vol. 10, no. 1, 2005.A. I. Legalov, “Functional language for creating architecturally independent parallel programs,” Computing technologies, vol. 10, no. 1, 2005.I. I. Levin, A. I. Dordopulo, I. V. Pisarenko, and A. K. Melnikov, “An approach to architecture-independent programming of computing systems based on the aspect-oriented Set@ language,” Proceedings of the Southern Federal University. Technical sciences. In Russian, vol. 197, no. 3, 2018.I. I. Levin, A. I. Dordopulo, I. V. Pisarenko, and A. K. Melnikov, “An approach to architecture-independent programming of computing systems based on the aspect-oriented Set@ language,” Proceedings of the Southern Federal University. Technical sciences. In Russian, vol. 197, no. 3, 2018.A. I. Legalov, V. S. Vasilev, I. V. Matkovskii, and M. S. Ushakova, “A toolkit for the development of data-driven functional parallel programmes,” in International Conference on Parallel Computational Technologies, 2018, pp. 16-30.A. I. Legalov, V. S. Vasilev, I. V. Matkovskii, and M. S. Ushakova, “A toolkit for the development of data-driven functional parallel programmes,” in International Conference on Parallel Computational Technologies, 2018, pp. 16-30.V. S. Vasilev and A. I. Legalov, “Loop-invariant Optimization in the Pifagor Language,” Automatic Control and Computer Sciences, vol. 52, no. 7, pp. 843-849, 2018.V. S. Vasilev and A. I. Legalov, “Loop-invariant Optimization in the Pifagor Language,” Automatic Control and Computer Sciences, vol. 52, no. 7, pp. 843-849, 2018.U. V. Udalova, A. I. Legalov, and N. U. Sirotinina, “Methods for debugging and verifying functional-stream parallel programs,” Journal of the Siberian Federal University. Equipment and technologies, vol. 4, no. 2, 2011.U. V. Udalova, A. I. Legalov, and N. U. Sirotinina, “Methods for debugging and verifying functional-stream parallel programs,” Journal of the Siberian Federal University. Equipment and technologies, vol. 4, no. 2, 2011.M. S. Ushakova and A. I. Legalov, “Verification of Programs with Mutual Recursion in Pifagor Language,” Automatic Control and Computer Sciences, vol. 52, no. 7, pp. 850-866, 2018.M. S. Ushakova and A. I. Legalov, “Verification of Programs with Mutual Recursion in Pifagor Language,” Automatic Control and Computer Sciences, vol. 52, no. 7, pp. 850-866, 2018.S. V. Zykov, Fundamentals of Modern Programming. Development of heterogeneous systems in an Internet-oriented environment. IPR Media, 2017.S. V. Zykov, Fundamentals of Modern Programming. Development of heterogeneous systems in an Internet-oriented environment. IPR Media, 2017.A. I. Legalov, I. A. Legalov, and I. V. Matkovsky, “Specifics of semantics of a statically typed language of functional and dataflow parallel programming,” in Scientific Conference Scientific Service on the Internet, 2019, pp. 489-500.A. I. Legalov, I. A. Legalov, and I. V. Matkovsky, “Specifics of semantics of a statically typed language of functional and dataflow parallel programming,” in Scientific Conference Scientific Service on the Internet, 2019, pp. 489-500.A. I. Legalov, I. V. Matkovsky, M. S. Ushakova, and D. S. Romanova, “Dynamically Changing Parallelism with the Asynchronous Sequential Data Flows,” Modeling and analysis of information systems, vol. 27, no. 2, pp. 164-179, 2020.A. I. Legalov, I. V. Matkovsky, M. S. Ushakova, and D. S. Romanova, “Dynamically Changing Parallelism with the Asynchronous Sequential Data Flows,” Modeling and analysis of information systems, vol. 27, no. 2, pp. 164-179, 2020.O. V. Nepomnyashchiy, I. N. Ryzhenko, and A. I. Legalov, “Method of architecture-independent high-level synthesis of VLSI,” Proceedings of the Southern Federal University. Technical sciences, vol. 202, no. 8, 2018.O. V. Nepomnyashchiy, I. N. Ryzhenko, and A. I. Legalov, “Method of architecture-independent high-level synthesis of VLSI,” Proceedings of the Southern Federal University. Technical sciences, vol. 202, no. 8, 2018.O. V. Nepomnyashchiy, I. N. Ryzhenko, and A. I. Legalov, “Methods, algorithms, and software tools for architecturally independent high-level synthesis of single-chip digital systems,” in Supercomputing Technologies (SCT-2018), 2018, pp. 104-109.O. V. Nepomnyashchiy, I. N. Ryzhenko, and A. I. Legalov, “Methods, algorithms, and software tools for architecturally independent high-level synthesis of single-chip digital systems,” in Supercomputing Technologies (SCT-2018), 2018, pp. 104-109.M. S. Ushakova, “Data type semantics of the dataflow parallel programming language Pifagor,” Educational resources and technologies, vol. 14, no. 2, 2016.M. S. Ushakova, “Data type semantics of the dataflow parallel programming language Pifagor,” Educational resources and technologies, vol. 14, no. 2, 2016.I. V. Matkovsky and A. I. Legalov, “Instrumental support for the translation and execution of functional-stream parallel programs,” Polzunovsky vestnik, no. 2, pp. 49-52, 2013.I. V. Matkovsky and A. I. Legalov, “Instrumental support for the translation and execution of functional-stream parallel programs,” Polzunovsky vestnik, no. 2, pp. 49-52, 2013.J. Ferrante, K. J. Ottenstein, and J. D. Warren, “The program dependence graph and its use in optimization,” ACM Transactions on Programming Languages and Systems (TOPLAS), vol. 9, no. 3, pp. 319-349, 1987.J. Ferrante, K. J. Ottenstein, and J. D. Warren, “The program dependence graph and its use in optimization,” ACM Transactions on Programming Languages and Systems (TOPLAS), vol. 9, no. 3, pp. 319-349, 1987.V. M. Bakanov, “Software tools for analyzing the information structure of algorithms based on their information graphs,” in Parallel Computing Technologies (PaVT ' 2016), 2016, pp. 432-441.V. M. Bakanov, “Software tools for analyzing the information structure of algorithms based on their information graphs,” in Parallel Computing Technologies (PaVT ' 2016), 2016, pp. 432-441.

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