A parallel time series algorithm for searching similar sub-sequences

<p><span>Dynamic time warping (DTW) is an important metric for measuring similarity most series applications. The computations of DTW cost too much especially with the gigantic sequence databases and lead to urgent need accelerating these computations. However, multi-core cluster systems, which are available now, their scalability performance/cost ratio, meet more powerful efficient performance. This paper proposes a highly parallel vectorized algorithm high performance computing DTW, addressed clusters using Intel quad-core Xeon co-processors. It deduces architecture. Implementations employ potential both message passing interface (MPI) OpenMP libraries. implementation based on programming technology offloads execution mode, where part code sub-sequences processor side, uploaded co-processor results experiments confirm effectiveness algorithm.</span></p>