Standard

High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System. / Belyaev, Nikolay; Perepelkin, Vladislav.

Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings. ed. / Victor Malyshkin. Springer Science and Business Media Deutschland GmbH, 2021. p. 143-150 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12942 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review

Harvard

Belyaev, N & Perepelkin, V 2021, High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System. in V Malyshkin (ed.), Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12942 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 143-150, 16th International Conference on Parallel Computing Technologies, PaCT 2021, Kaliningrad, Russian Federation, 13.09.2021. https://doi.org/10.1007/978-3-030-86359-3_11

APA

Belyaev, N., & Perepelkin, V. (2021). High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System. In V. Malyshkin (Ed.), Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings (pp. 143-150). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12942 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-86359-3_11

Vancouver

Belyaev N, Perepelkin V. High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System. In Malyshkin V, editor, Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 143-150. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-86359-3_11

Author

Belyaev, Nikolay ; Perepelkin, Vladislav. / High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System. Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings. editor / Victor Malyshkin. Springer Science and Business Media Deutschland GmbH, 2021. pp. 143-150 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

BibTeX

@inproceedings{d4a14c291e714fa78716cf129a6311b6,
title = "High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System",
abstract = "Automatic synthesis of efficient scientific parallel programs for supercomputers is in general a complex problem of system parallel programming. Therefore various specialized synthesis algorithms and heuristics are of use. LuNA system for automatic construction of distributed parallel programs provides a basis for accumulation of such algorithms to provide high-quality parallel programs generation in particular subject domains. If no specialized support is available in LuNA for given input, then the general synthesis algorithm is used, which does construct the required program, but its efficiency may be unsatisfactory. In the paper a specialized run-time system for LuNA is presented, which provides runtime support for dense linear algebra operations implementation on distributed memory multicomputers. Experimental results demonstrate, that automatically generated parallel programs of the class outperform corresponding ScaLAPACK library subroutines, which makes LuNA system practically applicable for generating high performance distributed parallel programs for supercomputers in the dense linear algebra application class.",
keywords = "Distributed dense linear algebra subroutines, Fragmented programming technology, LuNA system, Parallel programming automation",
author = "Nikolay Belyaev and Vladislav Perepelkin",
note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 16th International Conference on Parallel Computing Technologies, PaCT 2021 ; Conference date: 13-09-2021 Through 18-09-2021",
year = "2021",
doi = "10.1007/978-3-030-86359-3_11",
language = "English",
isbn = "9783030863586",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "143--150",
editor = "Victor Malyshkin",
booktitle = "Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings",
address = "Germany",

}

RIS

TY - GEN

T1 - High-Efficiency Specialized Support for Dense Linear Algebra Arithmetic in LuNA System

AU - Belyaev, Nikolay

AU - Perepelkin, Vladislav

N1 - Publisher Copyright: © 2021, Springer Nature Switzerland AG.

PY - 2021

Y1 - 2021

N2 - Automatic synthesis of efficient scientific parallel programs for supercomputers is in general a complex problem of system parallel programming. Therefore various specialized synthesis algorithms and heuristics are of use. LuNA system for automatic construction of distributed parallel programs provides a basis for accumulation of such algorithms to provide high-quality parallel programs generation in particular subject domains. If no specialized support is available in LuNA for given input, then the general synthesis algorithm is used, which does construct the required program, but its efficiency may be unsatisfactory. In the paper a specialized run-time system for LuNA is presented, which provides runtime support for dense linear algebra operations implementation on distributed memory multicomputers. Experimental results demonstrate, that automatically generated parallel programs of the class outperform corresponding ScaLAPACK library subroutines, which makes LuNA system practically applicable for generating high performance distributed parallel programs for supercomputers in the dense linear algebra application class.

AB - Automatic synthesis of efficient scientific parallel programs for supercomputers is in general a complex problem of system parallel programming. Therefore various specialized synthesis algorithms and heuristics are of use. LuNA system for automatic construction of distributed parallel programs provides a basis for accumulation of such algorithms to provide high-quality parallel programs generation in particular subject domains. If no specialized support is available in LuNA for given input, then the general synthesis algorithm is used, which does construct the required program, but its efficiency may be unsatisfactory. In the paper a specialized run-time system for LuNA is presented, which provides runtime support for dense linear algebra operations implementation on distributed memory multicomputers. Experimental results demonstrate, that automatically generated parallel programs of the class outperform corresponding ScaLAPACK library subroutines, which makes LuNA system practically applicable for generating high performance distributed parallel programs for supercomputers in the dense linear algebra application class.

KW - Distributed dense linear algebra subroutines

KW - Fragmented programming technology

KW - LuNA system

KW - Parallel programming automation

UR - http://www.scopus.com/inward/record.url?scp=85115321670&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-86359-3_11

DO - 10.1007/978-3-030-86359-3_11

M3 - Conference contribution

AN - SCOPUS:85115321670

SN - 9783030863586

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 143

EP - 150

BT - Parallel Computing Technologies - 16th International Conference, PaCT 2021, Proceedings

A2 - Malyshkin, Victor

PB - Springer Science and Business Media Deutschland GmbH

T2 - 16th International Conference on Parallel Computing Technologies, PaCT 2021

Y2 - 13 September 2021 through 18 September 2021

ER -

ID: 34337227