Geodesic
Mapping DVMH-programs with regular dependencies onto clusters with GPU
Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika, Tome 2 (2013) no. 4, pp. 44-56 Cet article a éte moissonné depuis la source Math-Net.Ru

Voir la notice de l'article

In the 2011 year DVMH programming model for new heterogeneous and hybrid supercomputer systems (or DVM for Heterogeneous systems) was introduced in the Keldysh Institute for Applied Mathematics of RAS. The developed high-level programming languages were based on standard Fortran and C programming languages, but extended with the directives for mapping the program onto a parallel computer. The directives are represented as special comments (or pragmas). The paper describes problems and methods for mapping loops, which have dependencies to the GPU. Efficiency of the developed Fortran DVMH parallel programs with regular dependencies is demonstrated.
Keywords: DVM for Heterogeneous systems, hybrid computational systems with accelerators, GPU, CUDA.
Mots-clés : Fortran DVMH 
@article{VYURV_2013_2_4_a3,
     author = {V. A. Bakhtin and A. S. Kolganov and V. A. Krukov and N. V. Podderyugina and M. N. Pritula},
     title = {Mapping {DVMH-programs} with regular dependencies onto clusters with {GPU}},
     journal = {Vestnik \^U\v{z}no-Uralʹskogo gosudarstvennogo universiteta. Seri\^a Vy\v{c}islitelʹna\^a matematika i informatika},
     pages = {44--56},
     year = {2013},
     volume = {2},
     number = {4},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VYURV_2013_2_4_a3/}
}
TY  - JOUR
AU  - V. A. Bakhtin
AU  - A. S. Kolganov
AU  - V. A. Krukov
AU  - N. V. Podderyugina
AU  - M. N. Pritula
TI  - Mapping DVMH-programs with regular dependencies onto clusters with GPU
JO  - Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika
PY  - 2013
SP  - 44
EP  - 56
VL  - 2
IS  - 4
UR  - http://geodesic.mathdoc.fr/item/VYURV_2013_2_4_a3/
LA  - ru
ID  - VYURV_2013_2_4_a3
ER  - 
%0 Journal Article
%A V. A. Bakhtin
%A A. S. Kolganov
%A V. A. Krukov
%A N. V. Podderyugina
%A M. N. Pritula
%T Mapping DVMH-programs with regular dependencies onto clusters with GPU
%J Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika
%D 2013
%P 44-56
%V 2
%N 4
%U http://geodesic.mathdoc.fr/item/VYURV_2013_2_4_a3/
%G ru
%F VYURV_2013_2_4_a3
V. A. Bakhtin; A. S. Kolganov; V. A. Krukov; N. V. Podderyugina; M. N. Pritula. Mapping DVMH-programs with regular dependencies onto clusters with GPU. Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika, Tome 2 (2013) no. 4, pp. 44-56. http://geodesic.mathdoc.fr/item/VYURV_2013_2_4_a3/

[1] Top500 List – November 2012 TOP500 Supercomputer Sites } {\tt http://top500.org/list/2012/11

[2] High Performance Fortran } {\tt http://hpff.rice.edu

[3] N. A. Konovalov, V. A. Krukov, A. A. Pogrebtsov, N. V. Podderyugina, Y. L. Sazanov, “Parallel programming in the DVM system. Fortran-DVM and C-DVM languages”, PACO’2001 (Moscow), Proceedings of International conference “Parallel computations and control problems”, 2001, 140–154

[4] N. A. Konovalov, V. A. Krukov, S. N. Mihailov, A. A. Pogrebtsov, “Fortran DVM – a language for mobile parallel programs development”, Programming, 1995, no. 1, 49–54

[5] N. A. Konovalov, V. A. Krukov, Y. L. Sazanov, “C-DVM – a language for mobile parallel programs development”, Programming, 1999, no. 1, 54–65

[6] R. Dolbeau, S. Bihan, F. Bodin., HMPPTM: A Hybrid Multi-core Parallel Programming Environment } {\tt http://www.caps-entreprise.com/wp-content/uploads/2012/08/capshmpp-gpgpu-Boston-Workshop-Oct-2007.pdf

[7] The Portland Group, PGI Accelerator Programming Model for Fortran C. } {\tt http://www.pgroup.com/lit/whitepapers/pgi_accel_prog_model_1.3.pdf

[8] OpenACC } {\tt http://www.openacc-standard.org

[9] T. D. Han, T. S. Abdelrahman, “hiCUDA: High-Level GPGPU Programming”, IEEE Transactions on Parallel and Distributed Systems, 22:3 (2011), 78-90

[10] Bakhtin V.A., Klinov M.S., Krukov V.A., Podderyugina N.V., Pritula M.N., Sazanov Y.L, “Extension of the DVM parallel programming model for clusters with heterogeneous nodes”, Vestnik Yuzhno-Ural’skogo gosudarstvennogo universiteta, seriya “Matematicheskoe modelirovanie i programmirovanie”, 18(277):12 (2012), 82–92

[11] S. J. Pennycook, S. D. Hammond, S. A. Jarvis, G. R. Mudalige, “Performance Analysis of a Hybrid MPI/CUDA Implementation of the NAS-LU Benchmark”, ACM SIGMETRICS Performance Evaluation Review – Special issue on the 1st international workshop on performance modeling, benchmarking and simulation of high performance computing systems (PMBS 10), 38:4 (2011), 23–29

[12] S. Seo, G. Jo, J. Lee, “Performance Characterization of the NAS Parallel Benchmarks in OpenCL”, 2011 IEEE International Symposium on. Workload Characterization (IISWC), 38:4 (2011), 137–148