Abstract
The False Nearest Neighbors (FNN) method is particularly relevant in several fields of science and engineering (medicine, economics, oceanography, biological systems, etc.). In some of these applications, it is important to give results within a reasonable time scale, so the execution time of the FNN method has to be reduced. This paper describes two parallel implementations of the FNN method based on the distribution of embedding dimensions for distributed memory architectures. A “Single-Program, Multiple Data” (SPMD) paradigm is employed using a simple data decomposition approach where each processor runs the same program but acts on a different subset of the data. The computationally intensive part of the method lies mainly in the neighbor search and this task is therefore parallelized and executed using 4 to 64 processors. The accuracy and performance of the two parallel approaches are then assessed and compared to the best sequential implementation of the FNN method which appears in the TISEAN project. The results indicate that the two parallel approaches, when the method is run using 64 processors on the MareNostrum supercomputer, are between 17 and 37 times faster than the sequential one. Efficiency is between 26% and 59%.
Similar content being viewed by others
References
Al-furaih I, Aluru S, Goil S, Ranka S (2000) Parallel construction of multidimensional binary search trees. IEEE Trans Parallel Distrib Syst 11:136–148
Arya S, Mount DM, Netanyahu NS, Silverman R, Wu AY (1998) An optimal algorithm for approximate nearest neighbor searching. J ACM 45:891–923
Asano T, Edahiro M, Imai H, Iri M, Murota K (1985) Practical use of bucketing techniques in computational geometry. In: Toussaint GT (ed) Computational geometry. Elsevier, Amsterdam, pp 153–195
Bentley JL (1990) K-d trees for semidynamic point sets. In: Proceedings of the sixth annual ACM symposium on computational geometry. ACM, New York, pp 187–197
De Berg M, Cheong O, Van Kreveld M, Overmars M (2008) Computational geometry: algorithms and applications. Springer, Berlin
Devroye L (1986) Lecture notes on bucket algorithms (Progress in computer science 6). Birkhäuser, Boston
Grama A, Gupta A, Karypis G, Kumar V (2003) Introduction to parallel computing. Pearson Education, Harlow
Grassberger P (1990) An optimized box-assisted algorithm for fractal dimensions. Phys Lett A 148:63–68
Grassberger P, Schreiber T, Schaffrath C (1991) Nonlinear time sequence analysis. Int J Bifurc Chaos 1:521–547
Gropp W, Lusk E, Skjellum A (1999) Using MPI: portable parallel programming with the message-passing interface. MIT Press, Cambridge
Hegger R, Kantz H, Schreiber T (1999) Practical implementation of nonlinear time series methods: The TISEAN package. Chaos 9:413–435
Kantz H, Schreiber T (2004) Nonlinear time series analysis. Cambridge University Press, Cambridge
Kennel MB, Brown R, Abarbanel HDI (1992) Determining embedding dimension for phase-space reconstruction using a geometrical construction. Phys Rev A 45:3403–3411
http://hpux.connect.org.uk/hppd/hpux/Physics/embedding-26.May.93
Kennel MB (2004) KDTREE Fortran 95 and C++ software to efficiently search for near neighbors in a multi-dimensional Euclidean space. arXiv:physics/0408067
Miller FP, Vandome AF, McBrewster J (2009) Kd-tree. VDM Publishing House, Beau Bassin
Moore A (1991) An introductory tutorial on kd-trees. Technical Report No. 209 (extract from PhD. Thesis), University of Cambridge
Packard NH, Crutchfield JP, Farmer JD, Shaw RS (1980) Geometry from a time series. Phys Rev Lett 45:712–716
Preparata FP, Shamos MI (1991) Computational geometry. An introduction. Springer, New York
Schreiber T (1995) Efficient neighbor searching in nonlinear time series analysis. Int J Bifurc Chaos 5:349–358
Sedgewick R (2001) Algorithms in C. Addison-Wesley, Reading
Takens F (1981) Detecting strange attractors in turbulence. In: Rand DA, Young L-S (eds) Dynamical systems and turbulence, Warwick 1980. Springer, New York, pp 366–381
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Marín Carrión, I., Arias Antúnez, E., Artigao Castillo, M.M. et al. A distributed memory architecture implementation of the False Nearest Neighbors method based on distribution of dimensions. J Supercomput 59, 1596–1618 (2012). https://doi.org/10.1007/s11227-011-0570-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-011-0570-z