Abstract
Modern computing clusters have complex interconnections between processors. The complete specification of a communication network is often insufficiently detailed. Methods for automatically refining specification are considered, which use algorithms for reconstructing the topology of connections between processors from results of benchmarking of the cluster communication network. Methods for 3D visualizing the cluster topology and for visually comparing the constructed topology with the template topology provided by the specification are also considered. The approach was tested on systems with the BlueGene P architecture and on the Lomonosov supercomputer. A greedy algorithm for retrieving the topology and an adaptation of the multidimensional scaling method for visualization are designed.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
L. Ventura, J. Antonio, H. Hacid, A. Ansiaux, and M. L. Maag, “Conversations reconstruction in the social web,” in Proc. 21st Int. Conf. Companion on World Wide Web, 2012 (ACM, New York, 2012), pp. 573–574.
R. Siamwalla, R. Sharma, and S. Keshav, Discovering Internet Topology. Technical Report (Cornell Univ. Comp. Sci. Department, 1907); http://www.cs.corenll.edu/skeshav/papers/discovery.pdf).
M. Coates, R. Castro, and R. Nowak, “Maximum likelihood network topology identification from edge-based unicast measurements,” in Proc. ACM SIGMETRICS Int. Conf. on Measurement and Modeling of Computer Systems, Marina del Rey, CA, June, 2002 (ACM, New York, 2002), pp. 11–20.
E. Hernandez-Lemus and J. M. Siqueiros-Garcia, “Information theoretical methods for complex network structure reconstruction,” Complex Adapt. Syst. Model. 1(1), 1–22 (2013).
C. Furtlehner, Y. Han, J.-M. Lasgouttes, V. Martin, and F. Moutarde, “Propagation of information on undirected dependency graphs for road traffic inference,” in Chaos, Complexity and Transport (Proc. CCT 11, Marseilles, France, May 23–27, 2011) (World Scientific Publishing, Singapore, 2012), pp. 191–203.
A. N. Salnikov, D. Yu. Andreev, and R. D. Lebedev, “Toolkit for Analyzing the Communication Environment Characteristics of a Computational Cluster Based on MPI Standard Functions,” Mos. Univ. Comput. Math. Cybern. 36, 41–49 (2012).
A. N. Salnikov, “PARUS: A paralle programming framework for heterogeneous multiprocessor systems,” in Proc. 15th Eur. PVM/MPI Users’ Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface. Lecture Notes in Computer Science. No. 4192 (Springer, Berlin, 2006), pp. 408–409.
I. Borg and P. J. F. Groenen, Springer Series in Statistics, Vol. XXII: Modern Multidimensional Scaling — Theory and Applications (Springer-Verlag, New York, 2005).
A. N. Salnikov and P. S. Bannikov, “A system for visualization of the results of MPI-benchmarking of interconnections in computer systems,” Vestn. Perm. Univ. Mat. Mekh. Informatika, No. 3, 80–85 (2012).
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © P.S. Bannikov, A.N. Salnikov, 2014, published in Vestnik Moskovskogo Universiteta. Vychislitel’naya Matematika i Kibernetika, 2014, No. 2, pp. 43–52.
About this article
Cite this article
Bannikov, P.S., Salnikov, A.N. Retrieving topology of interconnections in computational cluster based on results of MPI benchmarks. MoscowUniv.Comput.Math.Cybern. 38, 73–82 (2014). https://doi.org/10.3103/S0278641914020022
Received:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S0278641914020022