Source Code, Datasets

SING: A Data Series Similarity Search Index that Uses GPUs
Botao Peng, Panagiota Fatourou, Themis Palpanas



Data series similarity search is a core operation for several data series analysis applications across many different domains. This has attracted lots of interest that led to the development of several indexing techniques. In this work, we propose SING, the first data series index designed to take advantage of Graphics Processing Units (GPUs). SING is an in-memory index that uses the GPU's parallelization opportunities (as well as SIMD, multi-core and multi-socket), in order to accelerate similarity search. We conduct an experimental evaluation with several synthetic and real datasets, which shows that SING is up to 5.1x faster than the state-of-the-art parallel in-memory approach, and up to 62x faster than the state-of-the-art parallel serial scan algorithm. SING achieves exact similarity search query times as low as 32msec on 100GB datasets, which enables interactive data exploration on very large data series collections.


Source Code

You may freely use this code for research purposes, provided that you properly acknowledge the authors using the following reference:

Botao Peng, Panagiota Fatourou, Themis Palpanas. SING: Sequence Indexing Using GPUs. ICDE 2021.


Synthetic Datasets

We produced a set of synthetic datasets with sizes from 50 million to 200 million data series composed by random walks of length 256. Each data point in the data series is produced as xi+1=N(xi,1), where N(0,1) is a standard normal distribution.
The synthetic data generator code is included in the source code we are making available.


Real Datasets

Our method was tested on two real datasets.
  1. For our first real dataset, Seismic, we used the IRIS Seismic Data Access repository to gather data series representing seismic waves from various locations. We obtained 100 million data series of size 256. The complete dataset size was 100 GB.
  2. The second real dataset, SALD, includes neuroscience MRI data. The dataset comprised of 200 million data series of size 128. The complete dataset size was 100GB.