Geodesic
The use of MPI and OpenMP technologies for subsequence similarity search in very long time series on a computer cluster system with nodes based on the Intel Xeon Phi Knights Landing many-core processor
Numerical methods and programming, Tome 20 (2019) no. 1, pp. 29-44.

Voir la notice de l'article provenant de la source Math-Net.Ru

Nowadays, the subsequence similarity search is required in a wide range of time series mining applications: climate modeling, financial forecasts, medical research, etc. In most of these applications, the Dynamic Time Warping (DTW) similarity measure is used, since DTW is empirically confirmed as one of the best similarity measures for the majority of subject domains. Since the DTW measure has a quadratic computational complexity with respect to the length of query subsequence, a number of parallel algorithms for various many-core architectures are developed, namely FPGA, GPU, and Intel MIC. In this paper we propose a new parallel algorithm for subsequence similarity search in very large time series on computer cluster systems with nodes based on Intel Xeon Phi Knights Landing (KNL) many-core processors. Computations are parallelized on two levels as follows: by MPI at the level of all cluster nodes and by OpenMP within a single cluster node. The algorithm involves additional data structures and redundant computations, which make it possible to efficiently use the capabilities of vector computations on Phi KNL. Experimental evaluation of the algorithm on real-world and synthetic datasets shows that the proposed algorithm is highly scalable.
Keywords: time series, similarity search, parallel algorithm, OpenMP, Intel Xeon Phi, Knights Landing, data layout, vectorization. 
@article{VMP_2019_20_1_a3,
     author = {Ya. A. Kraeva and M. L. Tsymbler},
     title = {The use of {MPI} and {OpenMP} technologies for subsequence similarity search in very long time series on a computer cluster system with nodes based on the {Intel} {Xeon} {Phi} {Knights} {Landing} many-core processor},
     journal = {Numerical methods and programming},
     pages = {29--44},
     publisher = {mathdoc},
     volume = {20},
     number = {1},
     year = {2019},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2019_20_1_a3/}
}
TY  - JOUR
AU  - Ya. A. Kraeva
AU  - M. L. Tsymbler
TI  - The use of MPI and OpenMP technologies for subsequence similarity search in very long time series on a computer cluster system with nodes based on the Intel Xeon Phi Knights Landing many-core processor
JO  - Numerical methods and programming
PY  - 2019
SP  - 29
EP  - 44
VL  - 20
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2019_20_1_a3/
LA  - ru
ID  - VMP_2019_20_1_a3
ER  - 
%0 Journal Article
%A Ya. A. Kraeva
%A M. L. Tsymbler
%T The use of MPI and OpenMP technologies for subsequence similarity search in very long time series on a computer cluster system with nodes based on the Intel Xeon Phi Knights Landing many-core processor
%J Numerical methods and programming
%D 2019
%P 29-44
%V 20
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2019_20_1_a3/
%G ru
%F VMP_2019_20_1_a3
Ya. A. Kraeva; M. L. Tsymbler. The use of MPI and OpenMP technologies for subsequence similarity search in very long time series on a computer cluster system with nodes based on the Intel Xeon Phi Knights Landing many-core processor. Numerical methods and programming, Tome 20 (2019) no. 1, pp. 29-44. http://geodesic.mathdoc.fr/item/VMP_2019_20_1_a3/