Distributed and Parallel Delaunay Triangulation Construction with Balanced Binary-tree Model in Cloud

Delaunay triangulation (D-TIN) is an importantgraphic tool in computational geometry, which is not onlywidely used in many real applications, but also very significantfor many spatial data mining algorithms. However,constructing Delaunay triangulation is time-consuming formost practical applications. Distributed and parallel computingmechanism is becoming a good choice to solve large scale andcompute-intensive D-TIN applications. This paper proposes anovel hybrid algorithm (HA) for D-TIN construction in cloudcomputing environment, which is based on a balanced binarytreemodel and an elegant data structure called quad-edge. HAcombines the divide & conquer approach and the incrementalmethod. Moreover, a distributed and parallel version of Delaunaytriangulation computing service in cloud is designed andimplemented.The hybrid algorithm performed in both centralised andin cloud environments are compared. Experimental resultsshowed that the hybrid D-TIN service outperforms both thethe divide & conquer one and the incremental one, andit can effectively provide higher data mining services withfundamental D-TIN construction function in cloud

[1]  Yanfeng Zhang,et al.  iMapReduce: A Distributed Computing Framework for Iterative Computation , 2011, Journal of Grid Computing.

[2]  Jean-Daniel Boissonnat,et al.  Incremental construction of the delaunay triangulation and the delaunay graph in medium dimension , 2009, SCG '09.

[3]  Rajkumar Buyya,et al.  QoS-based Task Group Deployment on Grid by Learning the Performance Data , 2014, Journal of Grid Computing.

[4]  Jie Liu,et al.  A new compound algorithm study for Delaunay triangulation construction , 2007, Geoinformatics.

[5]  Chen Chongcheng,et al.  Distributed Spatial Data Mining in Geospatial Knowledge Service Grid , 2010, 2010 Second International Conference on Advanced Geographic Information Systems, Applications, and Services.

[6]  Wolfgang Mulzer,et al.  Delaunay Triangulations in O(sort(n)) Time and More , 2009, 2009 50th Annual IEEE Symposium on Foundations of Computer Science.

[7]  Kieran F. Mulchrone.  Application of Delaunay triangulation to the nearest neighbour method of strain analysis , 2003 .

[8]  Olivier Devillers.  Improved incremental randomized Delaunay triangulation , 1998, SCG '98.

[9]  Leonidas J. Guibas,et al.  Primitives for the manipulation of general subdivisions and the computation of Voronoi diagrams , 1983, STOC.

[10]  Mark de Berg,et al.  Computational geometry: algorithms and applications , 1997 .

[11]  Chongcheng Chen,et al.  Parallel and Distributed Spatial Outlier Mining in Grid: Algorithm, Design and Application , 2015, Journal of Grid Computing.

[12]  Chen Chongcheng,et al.  Research on grid based spatial outliers mining and its implementation , 2008, 2008 3rd International Conference on Intelligent System and Knowledge Engineering.

[13]  Leonidas J. Guibas,et al.  Primitives for the manipulation of general subdivisions and the computation of Voronoi diagrams , 1983, STOC.

[14]  Howard J. Hamilton,et al.  Density-Based Spatial Clustering in the Presence of Obstacles and Facilitators , 2004, PKDD.

[15]  M. Bikdash,et al.  Algorithm to locate points in a Delaunay triangulation , 2006, 2006 Proceeding of the Thirty-Eighth Southeastern Symposium on System Theory.

[16]  Leonidas J. Guibas,et al.  Randomized incremental construction of Delaunay and Voronoi diagrams , 1990, Algorithmica.

[17]  C. Gold,et al.  Delete and insert operations in Voronoi/Delaunay methods and applications , 2003 .

[18]  Jeffrey W. Ohlmann,et al.  A randomized Delaunay triangulation heuristic for the Euclidean Steiner tree problem in ℜd , 2011, J. Heuristics.

[19]  Xianfeng Li,et al.  A new large-scale terrain real-time simplification algorithm based on Delaunay triangulation , 2007, Other Conferences.