Hybrid MPI/Open-MP acceleration approach for high-order schemes for CFD

• Autorzy: Saczek Michał , Wawrzak Karol , Tyliszczak Artur , Boguslawski Andrzej

Identyfikatory

DOI

10.17466/tq2018/22.3/e

• Języki publikacji: EN

Abstrakty

EN

The paper presents a hybridMPI+OpenMP (Message Passing Interface/Open Multi-Processor) algorithm used for parallel programs based on the high-order compact method.The main tools used to implement parallelism in computations are OpenMP andMPI whichdiffer in terms of memory on which they are based. OpenMP works on shared-memory and the MPIon distributed-memory whereas the hybrid model is based on a combination of those methods. The tests performed and described in this paper present significant advantages provided by a combination of the MPI/OpenMP approach. The test computations needed for verifying possibilities ofMPI, Open-MP and Hybrid of both tools were carried out using anacademic high-order SAILORsolver. The obtained results seem to be very promising to accelerate simulations of fluid flows as well as for application using high order methods.

Słowa kluczowe

EN

MPI; OpenMP; hybrid MPI/OpenMP; high-order methods

• Wydawca: Politechnika Gdańska
• Czasopismo: TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk
• Rocznik: 2018
• Tom: Vol. 22, No 3
• Strony: 179--193
• Opis fizyczny: Bibliogr. 24 poz., rys., tab.

Twórcy

autor

Saczek Michał

• Czestochowa University of Technology, Faculty of Mechanical Engineering and Computer Science, Armii Krajowej 21, 42-201 Czestochowa, Poland

autor

Wawrzak Karol

• Czestochowa University of Technology, Faculty of Mechanical Engineering and Computer Science, Armii Krajowej 21, 42-201 Czestochowa, Poland

autor

Tyliszczak Artur

• Czestochowa University of Technology, Faculty of Mechanical Engineering and Computer Science, Armii Krajowej 21, 42-201 Czestochowa, Poland

autor

Boguslawski Andrzej

• Czestochowa University of Technology, Faculty of Mechanical Engineering and Computer Science, Armii Krajowej 21, 42-201 Czestochowa, Poland

Bibliografia

• [1] Wang Z, Fidkowski K, Abgrall R, Bassi F, Caraeni D, Cary A, Deconinck H, Hartmann R,Hillewaert K, Huynh H, Kroll N, May G, Persson P, Leer B and Visbal M 2013 Int. J.Numer. Methods Fluids 72 811
• [2] Lele S 1992 J. Comput. Phys.103 16
• [3] Hockney R 1965 J.ACM12 16
• [4] Wang H 1981 ACMTrans. Math. Softw.7 170
• [5] Mattor N, Williams T and Hewett D 1995 Parallel Comput. 21 1769
• [6] Sun H, Zhang H and Ni L 1992 IEEEon Trans.Comput. 41 286
• [7] Belov P, Nugumanov E and Yakovlev S 2015 Preprint, arXiv 1505.06864
• [8] Stone H 1975 ACMTrans.Math. Softw.1 289
• [9] Rao S C S 2008 Parallel Comput.34 177
• [10] Qin J and Nguyen D 1998 Adv. Eng. Softw. 29 395
• [11] Afzal A, Ansari Z, Faizabadi A and Ramis M 2017 Arch. Comput. Method E 24 337
• [12] Dagum Land Menon R 1998 IEEEComput. Sci. Eng.5 46
• [13] William D, Lusk E and Skjellum A 1999 UsingMPI: Portable Parallel Programming withthe Message-Passing Interface 2nd edition,MIT Press, Cambridge
• [14] Rabenseifner R, Hager G and Jost G 2009 Proc. Euromicro Int. Conf. Parallel Distrib.Netw. Based Process 17 427
• [15] Amritkar A, Deb S and Tafti D 2014 J. Comput. Phys.25 6501
• [16] Mavriplis D 2002 Int. J. High Perform. Comput. Appl.16 395
• [17] Maknickas A, Kačeniauskas A, Kačianauskas R, Balevičius R and Algis Džiugys A 2006 Informatic a17 207
• [18] Jia R and Sundén B 2004 Comput. Fluids 33 57
• [19] Mininni P, Rosenberg D, Reddy R and Pouquet A 2011 Parallel Comput. 37 316
• [20] Gropp W, Kaushik D, Keyes D and Smith B 2001 Parallel Comput. 27 337
• [21] Wawrzak K, Boguslawski A and Tyliszczak A 2015 Flow Turbul. Combust. 95 437
• [22] Aniszewski W, Boguslawski A, Marek M and Tyliszczak A 2012 J. Comput. Phys. 231 7368
• [23] Wawrzak A and Tyliszczak A 2017 Arch. Mech .69 157
• [24] Tyliszczak A 2014 J. Comput. Phys. 276 438

Uwagi

PL

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2019).