Geodesic
A parallel discord discovery algorithm for time series on many-core accelerators
Numerical methods and programming, Tome 20 (2019) no. 3, pp. 211-223.

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

Discord is a refinement of the concept of anomalous subsequence of a time series. The discord discovery problem frequently occurs in a wide range of application areas related to time series: medicine, economics, climate modeling, etc. In this paper we propose a new parallel discord discovery algorithm for many-core systems in the case when the input data fit in the main memory. The algorithm exploits the ability to independently calculate the Euclidean distances between the subsequences of the time series. Computations are paralleled using OpenMP and OpenAcc for the Intel MIC (Many Integrated Core) and NVIDIA GPU platforms, respectively. The algorithm consists of two stages, namely precomputations and discovery. At the precomputation stage, we construct the auxiliary matrix data structures to ensure the efficient vectorization of computations on an accelerator. At the discovery stage, the algorithm searches for a discord based on the constructed structures. A number of numerical experiments confirm a high scalability of the proposed algorithm.
Keywords: time series, discord discovery, parallel algorithm, vectorization, OpenMP, OpenAcc, Intel Xeon Phi, NVIDIA GPU. 
@article{VMP_2019_20_3_a3,
     author = {M. L. Tsymbler},
     title = {A parallel discord discovery algorithm for time series on many-core accelerators},
     journal = {Numerical methods and programming},
     pages = {211--223},
     publisher = {mathdoc},
     volume = {20},
     number = {3},
     year = {2019},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2019_20_3_a3/}
}
TY  - JOUR
AU  - M. L. Tsymbler
TI  - A parallel discord discovery algorithm for time series on many-core accelerators
JO  - Numerical methods and programming
PY  - 2019
SP  - 211
EP  - 223
VL  - 20
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2019_20_3_a3/
LA  - ru
ID  - VMP_2019_20_3_a3
ER  - 
%0 Journal Article
%A M. L. Tsymbler
%T A parallel discord discovery algorithm for time series on many-core accelerators
%J Numerical methods and programming
%D 2019
%P 211-223
%V 20
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2019_20_3_a3/
%G ru
%F VMP_2019_20_3_a3
M. L. Tsymbler. A parallel discord discovery algorithm for time series on many-core accelerators. Numerical methods and programming, Tome 20 (2019) no. 3, pp. 211-223. http://geodesic.mathdoc.fr/item/VMP_2019_20_3_a3/