Advertisement

Programming and Computer Software

, Volume 45, Issue 3, pp 121–132 | Cite as

DVM-Approach to the Automation of the Development of Parallel Programs for Clusters

  • V. A. BakhtinEmail author
  • V. A. KrukovEmail author
Article
  • 9 Downloads

Abstract

The DVM-approach to the development of parallel programs for heterogeneous computer clusters with accelerators. The basic capabilities of DVM and SAPFOR , which automate the parallelization of applications, are discussed.

Notes

REFERENCES

  1. 1.
    Eisymont, L.L., Hybrid strategy of the development of hardware components, Otkrytye sist., 2017, no, 2, pp.8–11. https://www.osp.ru/os/2017/02/13052216Google Scholar
  2. 2.
    Top500 List - November 2018. http://top500.org/list/ 2018/11/Google Scholar
  3. 3.
    OpenMP 4.0 Specifications. https://www.openmp.org/specifications/Google Scholar
  4. 4.
    OpenACC Specification. https://www.openacc.org/ specificationGoogle Scholar
  5. 5.
    Bakhtin, V.A., Klinov, M.S., Krukov, V.A., Podderyugina, N.V., Pritula, M.N., and Sazanov, Yu.L., Extension of the DVM parallel programming model for clusters with heterogeneous nodes, Vestn. Yuzhno-Ural’sk. Gos. Univ., Ser. Mat. Model. Program., 2012, no. 18, pp. 82–92.Google Scholar
  6. 6.
    Bakhtin, V.A., Klinov, M.S., Kolganov, A.S., Krukov, V.A., Podderyugina, N.V., and Pritula, M.N., Automatic mapping of Fortran programs to clusters with GPUs, Vestn. Yuzhno-Ural’sk. Gos. Univ., Ser. Vychisl. Mat. Inform., 2014, vol. 3, no. 3, pp. 86–96.Google Scholar
  7. 7.
    DVM—the system for developing parallel programs. http://dvm-system.org/Google Scholar
  8. 8.
    Fortran-DVMH, Fortran-DVMH compiler, compilation, execution, and debugging DVMH programs. http: //dvm-system.org/static_data/tutorial/FDVMH-tutorial.pdfGoogle Scholar
  9. 9.
    C-DVMH, C-DVMH compilation, execution, and debugging DVMH programs. http://dvm-system.org /static_data/docs/CDVMH-reference-ru.pdfGoogle Scholar
  10. 10.
    Bakhtin, V.A., Kolganov, A.S., Krukov, V.A., Podderyugina, N.V., and Pritula, M.N., Methods of the dynamic configuration of DVMH programs for clusters with accelerators, Trudy mezhdunarodnoi konferentsii Superkom’yuternye dni v Rossii (Int. Conf. Supercomputer days in Russia), Moscow, 2015, Moscow: Mosk. Gos. Univ., 2015, pp. 257–268.Google Scholar
  11. 11.
    Kryukov, V.A. and Udovichenko, R.V., Debugging DVM programs, Program. Comput. Software, 2001, vol. 27, no. 3, pp. 123–129.CrossRefzbMATHGoogle Scholar
  12. 12.
    Krukov, V.A. and Kudryavtsev, M.V., Automating the debugging of parallel programs, Vychisl. Met. Program., 2006, vol. 7, no. 4, pp. 102–110.Google Scholar
  13. 13.
    Ermichev, A. A. and Krukov, V.A., Elaboration of the comparative technique for debugging DVMH programs, Trudy XIX vserossiiskoi konferentsii Nauchnyi servis v ceti Internet (Proc.of the All-Russia Conf. Scientific Services on the Internet), Novorossiisk, 2017, Moscow, Inst. Prikl. Mat., 2017, pp. 150–156.Google Scholar
  14. 14.
    Bakhtin, V.A., Zhukova, O.F., Kataev, N.A., Kolganov, A.S., Krukov, V.A., Kuznetsov, M.Yu., Podderyugina, N.V., and Pritula, M.N., Savitskaya, O.A., and Smirnov, A.A., Parallelization of Software Packages: Problems and Prospects, Trudy XIX vserossiiskoi konferentsii Nauchnyi servis v ceti Internet (Proc.of the All-Russia Conf. Scientific Services on the Internet), Novorossiisk, 2018, Moscow, Inst. Prikl. Mat., 2018, pp. 63–72.Google Scholar
  15. 15.
    Kataev, N.A., Kolganov, A.S., and Titov, P.A., the problem of simulating the propagation of elastic waves in media with complex 3D surface geometry to clusters of various architectures, Trudy XI mezhdunarodnoi konferentsii Parallel’nye vychislitel’nye tekhnologii (Proc. of the XI Int. Conf. on Parallel Computation Technologies), Kazan’, 2017, Chelyabinsk, Uzhno-Ural’skii Gos. Univ., 2017, pp. 341–355.Google Scholar
  16. 16.
    Bakhtin, V.A., Klinov, M.S., Krukov, V.A., and Podderyugina, N.V., Automatic parallelization of sequential programs for multicore clusters, Trudy mezhdunarodnoi superkomp’yuternoi konferentsii (Proc.of the Int. Conf. on Supercomputers), in Nauchnyi servis v ceti Internet (Scientific Services on the Internet), Ross. Akad. Nauk, 2010, pp. 12–15.Google Scholar
  17. 17.
    Aleksakhin, V.F., Bakhtin, V.A., Zakharov, A.S., Kolganov, A.S., Korolev, A.V., Krukov, V.A., Podderyugina, N.V., and Pritula, M.N., Experience of solving applied problems using the DVM system, Trudy mezhdunarodnoi konferentsii Superkom’yuternye dni v Rossii (Int. Conf. Supercomputer days in Russia), Moscow, 2017, Moscow: Mosk. Gos. Univ., 2017, pp. 650–661.Google Scholar
  18. 18.
    Bakhtin, V.A., Krukov, V.A., Chetverushkin, B.N., and Shil’nikov, E.V., Extension of the DVM parallel programming model for clusters with heterogeneous nodes, Doklady Math., 2011, vol. 84, no. 3, pp. 879–881.MathSciNetCrossRefGoogle Scholar
  19. 19.
    Bakhtin, V.A., Klinov, M.S., Kolganov, A.S., Krukov, V.A., Podderyugina, N.V., and Pritula, M.N., Automatic mapping of Fortran programs to clusters with accelerators, Trudy mezhdunarodnoi superkomp’yuternoi konferentsii (Proc.of the Int. Conf. on Supercomputers), in Nauchnyi servis v ceti Internet: mnogoobrazie superkomp’yuternykh mirov (Scientific Services on the Internet: the Diversity of Supercomputer Worlds), Ross. Akad. Nauk, 2014, pp. 17–22.Google Scholar
  20. 20.
    Pritula, M.N., Mapping DVMH programs to clusters with GPUs, Cand. Sci. (Phys.-Math.) Dissertation, Moscow: Keldysh Institute of Applied Mathematics, 2013, p. 105. http://keldysh.ru/council/1/pritula-diss.pdfGoogle Scholar
  21. 21.
    Voronkov, A.V., Golubev, A.S., Krukov, V.A., and Sychugova, E.P., Parallel version of the package REAKTOR-GP, Vopr. At. Nauki Tekh., Ser. Obesp. Bezopasnosti AES, 2009, no. 4, pp. 64–74.Google Scholar
  22. 22.
    Aleksakhin, V.F., Bakhtin, V.A., Zhukova, O.F., Kolganov, A.S., Krukov, V.A., Podderyugina, N.V., Pritula, M.N., Savitskaya, O.A., and Shubert, A.V., Parallelization of Fortran DVMH NAS NPB3.3.1 benchmarks for GPUs, Vestn. Ufa Gos. Aviatsionno-tekhnich. Inst., 2015, vol. 19, no. 1 (67), pp. 240–250.Google Scholar
  23. 23.
    Aleksakhin, V.F., Bakhtin, V.A., Zhukova, O.F., Kolganov, A.S., Krukov, V.A., Ostrovskaya, I.P., Podderyugina, N.V., Pritula, M.N., and Savitskaya, O.A., Parallelization of Fortran DVMH NAS NPB3.3.1 benchmarks for the coprocessor Intel Xeon Phi, Trudy mezhdunarodnoi konferentsii Parallel’nye vychislitel’nye tekhnologii (Proc. of the Int. Conf. on Parallel Computation Technologies), Chelyabinsk, Uzhno-Ural’skii Gos. Univ., 2015, pp. 19–30.Google Scholar
  24. 24.
    Krukov, V.A., Development of parallel programs for computer clusters and networks, Inf. tekhnolg. komput. sist., 2003, no. 1–2, pp. 42–61. ftp://ftp.keldysh.ru/ dvm-distr/journ1-2_page42_61.pdfGoogle Scholar
  25. 25.
    The efficiency of executing NAS NPB benchmarks. http://dvm-system.org/ru/category/performance/Google Scholar

Copyright information

© Pleiades Publishing, Ltd. 2019

Authors and Affiliations

  1. 1.Keldysh Institute of Applied Mathematics, Russian Academy of SciencesMoscowRussia
  2. 2.Moscow State UniversityMoscowRussia

Personalised recommendations