Approach to Accelerating Dissolved Vector Buffer Generation in Distributed In-Memory Cluster Architecture
AbstractThe buffer generation algorithm is a fundamental function in GIS, identifying areas of a given distance surrounding geographic features. Past research largely focused on buffer generation algorithms generated in a stand-alone environment. Moreover, dissolved buffer generation is data- and computing-intensive. In this scenario, the improvement in the stand-alone environment is limited when considering large-scale mass vector data. Nevertheless, recent parallel dissolved vector buffer algorithms suffer from scalability problems, leaving room for further optimization. At present, the prevailing in-memory cluster-computing framework—Spark—provides promising efficiency for computing-intensive analysis; however, it has seldom been researched for buffer analysis. On this basis, we propose a cluster-computing-oriented parallel dissolved vector buffer generating algorithm, called the HPBM, that contains a Hilbert-space-filling-curve-based data partition method, a data skew and cross-boundary objects processing strategy, and a depth-given tree-like merging method. Experiments are conducted in both stand-alone and cluster environments using real-world vector data that include points and roads. Compared with some existing parallel buffer algorithms, as well as various popular GIS software, the HPBM achieves a performance gain of more than 50%. View Full-Text
Share & Cite This Article
Shen, J.; Chen, L.; Wu, Y.; Jing, N. Approach to Accelerating Dissolved Vector Buffer Generation in Distributed In-Memory Cluster Architecture. ISPRS Int. J. Geo-Inf. 2018, 7, 26.
Shen J, Chen L, Wu Y, Jing N. Approach to Accelerating Dissolved Vector Buffer Generation in Distributed In-Memory Cluster Architecture. ISPRS International Journal of Geo-Information. 2018; 7(1):26.Chicago/Turabian Style
Shen, Jinxin; Chen, Luo; Wu, Ye; Jing, Ning. 2018. "Approach to Accelerating Dissolved Vector Buffer Generation in Distributed In-Memory Cluster Architecture." ISPRS Int. J. Geo-Inf. 7, no. 1: 26.
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.