Geodesic
Optimizing processes mapping for tasks with non-uniform data exchange run on cluster with different interconnects
Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika, Tome 4 (2015) no. 2, pp. 5-19 Cet article a éte moissonné depuis la source Math-Net.Ru

Voir la notice de l'article

The problem of mapping the parallel task to the nodes of computing cluster is considered. MPI software with non-uniform communication and heterogeneous interconnect of computing cluster require to appropriate parallel processes mapping for optimization of data exchange. The graph mapping algorithm is developed. It uses parallel program representation as a task graph and cluster topology representation as system graph. The proposed optimization technique is tested on synthetic benchmark and on real QBox software to study its efficiency on large number of computing cores. The positive results of optimization are achieved and the summary is presented in the paper. Speedup of 17-20
Keywords: task mapping, cluster, MPI.
Mots-clés : communication graph 
@article{VYURV_2015_4_2_a0,
     author = {V. V. Getmanskiy and V. S. Chalyshev and D. I. Kryzhanovskiy and E. I. Leksikov},
     title = {Optimizing processes mapping for tasks with non-uniform data exchange run on cluster with different interconnects},
     journal = {Vestnik \^U\v{z}no-Uralʹskogo gosudarstvennogo universiteta. Seri\^a Vy\v{c}islitelʹna\^a matematika i informatika},
     pages = {5--19},
     year = {2015},
     volume = {4},
     number = {2},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VYURV_2015_4_2_a0/}
}
TY  - JOUR
AU  - V. V. Getmanskiy
AU  - V. S. Chalyshev
AU  - D. I. Kryzhanovskiy
AU  - E. I. Leksikov
TI  - Optimizing processes mapping for tasks with non-uniform data exchange run on cluster with different interconnects
JO  - Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika
PY  - 2015
SP  - 5
EP  - 19
VL  - 4
IS  - 2
UR  - http://geodesic.mathdoc.fr/item/VYURV_2015_4_2_a0/
LA  - ru
ID  - VYURV_2015_4_2_a0
ER  - 
%0 Journal Article
%A V. V. Getmanskiy
%A V. S. Chalyshev
%A D. I. Kryzhanovskiy
%A E. I. Leksikov
%T Optimizing processes mapping for tasks with non-uniform data exchange run on cluster with different interconnects
%J Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika
%D 2015
%P 5-19
%V 4
%N 2
%U http://geodesic.mathdoc.fr/item/VYURV_2015_4_2_a0/
%G ru
%F VYURV_2015_4_2_a0
V. V. Getmanskiy; V. S. Chalyshev; D. I. Kryzhanovskiy; E. I. Leksikov. Optimizing processes mapping for tasks with non-uniform data exchange run on cluster with different interconnects. Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika, Tome 4 (2015) no. 2, pp. 5-19. http://geodesic.mathdoc.fr/item/VYURV_2015_4_2_a0/

[1] Kopysov S.P., Novikov A.K., Tonkov L.E., et al., “Methods of Parallel Processes and Threads Binding to Multicore Cluster Nodes”, Bulletin of Udmurt University: Mathematics, Mechanics and Computing Science, 2010, no. 1, 123–132 | MR

[2] Kurnosov M.G., “Mapping of parallel program branches to computing cores in distributed system”, Multiprocessor computing and control systems (Divnomorskoe, Gelendzhik, Russia, 2007), v. 1, 2007, 227–231

[3] C. Karlsson, T. Davies, Z. Chen, “Optimizing Process-to-Core Mappings for Application Level Multidimensional MPI Communications”, Cluster Computing (CLUSTER), IEEE International Conf. Proceedings (Beijing, China, September, 24-28, 2012), Beijing, 2012, 486–494 | DOI

[4] J. Zhang, J. Zhai, W. Chen, et al., “Process Mapping for MPI Collective Communications”, Lecture Notes in Computer Science, 5704, 2009, 81–92 | DOI

[5] H. Chen, W. Chen, J. Huang, et al., “MPIPP: an Automatic Profile-Guided Parallel Process Placement Toolset for SMP Clusters and Multiclusters”, Proceedings of the 20th annual international conference on Supercomputing, ICS'06 (Queensland, Australia, June, 28 - July, 01, 2006), Queensland, 2006, 353–360 | DOI | Zbl

[6] Intel\circledR MPI Library Reference Manual, (data obrascheniya: 20.12.2014) http://software.intel.com/sites/products/documentation/hpc/ics/impi/41/lin/Reference_Manual/index.htm

[7] P. Larsson, Shared Memory Communication vs. Infiniband, (data obrascheniya: 20.12.2014) http://www.nsc.liu.se/ pla/blog/2013/09/12/smp-vs-infiniband

[8] F. Gygi, R.K. Yates, J. Lorenz, et al., “Large-Scale First-Principles Molecular Dynamics Simulations on the Blue-Gene/L Platform using the Qbox Code”, Proceedings of the ACM/IEEE SC 2005 Conference (Seattle, WA, USA, November, 12-18, 2005), Seattle, 2005, 24 | DOI

[9] Tornado SUSU Supercomputer, (accessed: 20.12.2014) http://supercomputer.susu.ac.ru/en/computers/tornado