Scalable Algorithm for Subsequence Similarity Search in Very Large Time Series Data on Cluster of Phi KNL

Abstract

Nowadays, subsequence similarity search under the Dynamic Time Warping (DTW) similarity measure is applied in a wide range of time series mining applications. Since the DTW measure has a quadratic computational complexity w.r.t. the length of query subsequence, a number of parallel algorithms for various many-core architectures have been developed, namely FPGA, GPU, and Intel MIC. In this paper, we propose a novel parallel algorithm for subsequence similarity search in very large time series data on computing cluster with nodes based on the Intel Xeon Phi Knights Landing (KNL) many-core processors. Computations are parallelized both at the level of all cluster nodes through MPI, and within a single cluster node through OpenMP. The algorithm involves additional data structures and redundant computations, which make it possible to effectively use Phi KNL for vector computations. Experimental evaluation of the algorithm on real-world and synthetic datasets shows that it is highly scalable.