Technical Report: Towards Efficient Indexing of Spatiotemporal Trajectories on the GPU for Distance Threshold Similarity Searches

Applications in many domains require processing moving object trajectories. In this work, we focus on a trajectory similarity search that finds all trajectories within a given distance of a query trajectory over a time interval, which we call the distance threshold similarity search. We develop three indexing strategies with spatial, temporal and spatiotemporal selectivity for the GPU that differ significantly from indexes suitable for the CPU, and show the conditions under which each index achieves good performance. Furthermore, we show that the GPU implementations outperform multithreaded CPU implementations in a range of experimental scenarios, making the GPU an attractive technology for processing moving object trajectories. We test our implementations on two synthetic and one real-world dataset of a galaxy merger.

[1]  Timos K. Sellis,et al.  Spatio-temporal indexing for large multimedia applications , 1996, Proceedings of the Third IEEE International Conference on Multimedia Computing and Systems.

[2]  Le Gruenwald,et al.  Parallel online spatial and temporal aggregations on multi-core CPUs and many-core GPUs , 2014, Inf. Syst..

[3]  Tikara Hosino,et al.  Multi‐GPU algorithm for k‐nearest neighbor problem , 2012, Concurr. Comput. Pract. Exp..

[4]  G. Lake,et al.  Habitability in different Milky Way stellar environments: a stellar interaction dynamical approach. , 2013, Astrobiology.

[5]  Michael Gowanlock.  In-Memory Distance Threshold Queries on Moving Object Trajectories , 2014 .

[6]  Chris Jermaine,et al.  Closest-Point-of-Approach Join for Moving Object Histories , 2006, 22nd International Conference on Data Engineering (ICDE'06).

[7]  Tianyi David Han,et al.  Reducing branch divergence in GPU programs , 2011, GPGPU-4.

[8]  Le Gruenwald,et al.  U2STRA: high-performance data management of ubiquitous urban sensing trajectories on GPGPUs , 2012, CDMW '12.

[9]  Marios Hadjieleftheriou,et al.  R-Trees - A Dynamic Index Structure for Spatial Searching , 2008, ACM SIGSPATIAL International Workshop on Advances in Geographic Information Systems.

[10]  Martin Krulis,et al.  Combining CPU and GPU architectures for fast similarity search , 2012, Distributed and Parallel Databases.

[11]  M. Gowanlock,et al.  A model of habitability within the Milky Way galaxy. , 2010, Astrobiology.

[12]  Jörg Sander,et al.  A Trajectory Splitting Model for Efficient Spatio-Temporal Indexing , 2005, VLDB.

[13]  Nikos Pelekis,et al.  Nearest Neighbor Search on Moving Object Trajectories , 2005, SSTD.

[14]  Dinesh Manocha,et al.  Fast GPU-based locality sensitive hashing for k-nearest neighbor computation , 2011, GIS.

[15]  Dino Pedreschi,et al.  Trajectory pattern mining , 2007, KDD '07.

[16]  Suzanne L. Hawley,et al.  The Palomar/MSU Nearby Star Spectroscopic Survey. IV. The Luminosity Function in the Solar Neighborhood and M Dwarf Kinematics , 2002 .

[17]  Ralf Hartmut Güting,et al.  Efficient k-nearest neighbor search on moving object trajectories , 2010, The VLDB Journal.

[18]  Le Gruenwald,et al.  Parallel spatial query processing on GPUs using R-trees , 2013, BigSpatial '13.

[19]  Henri Casanova,et al.  Parallel In-Memory Distance Threshold Queries on Trajectory Databases , 2014, DBKDA 2014.

[20]  Nikos Pelekis,et al.  Algorithms for Nearest Neighbor Search on Moving Object Trajectories , 2007, GeoInformatica.

[21]  Henri Casanova,et al.  Distance threshold similarity searches on spatiotemporal trajectories using GPGPU , 2014, 2014 21st International Conference on High Performance Computing (HiPC).

[22]  Ralf Hartmut Güting,et al.  A data model and data structures for moving objects databases , 2000, SIGMOD '00.

[23]  Martin D. F. Wong,et al.  Parallel implementation of R-trees on the GPU , 2012, 17th Asia and South Pacific Design Automation Conference.

[24]  Jignesh M. Patel,et al.  Indexing Large Trajectory Data Sets With SETI , 2003, CIDR.

[25]  Petko Bakalov,et al.  On-line discovery of flock patterns in spatio-temporal data , 2009, GIS.

[26]  Samuel Madden,et al.  TrajStore: An adaptive storage system for very large trajectory data sets , 2010, 2010 IEEE 26th International Conference on Data Engineering (ICDE 2010).

[27]  Jae-Gil Lee,et al.  MoveMine: mining moving object databases , 2010, SIGMOD Conference.

[28]  Piet Hut,et al.  A hierarchical O(N log N) force-calculation algorithm , 1986, Nature.

[29]  Dimitrios Gunopulos,et al.  Efficient Indexing of Spatiotemporal Objects , 2002, EDBT.

[30]  Yunjun Gao,et al.  Efficient k-Nearest-Neighbor Search Algorithms for Historical Moving Object Trajectories , 2007, Journal of Computer Science and Technology.

[31]  N. Gehrels,et al.  Gamma-Ray Bursts and the Earth: Exploration of Atmospheric, Biological, Climatic, and Biogeochemical Effects , 2005, astro-ph/0505472.

[32]  Christian S. Jensen,et al.  Discovery of convoys in trajectory databases , 2008, Proc. VLDB Endow..