The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Articles | Volume XLIII-B4-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B4-2022, 639–644, 2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B4-2022, 639–644, 2022
02 Jun 2022
02 Jun 2022


H. Zhang1,2, M. Du1, W. Huang2, L. Ding2, D. Tang2, and J. Jiang1 H. Zhang et al.
  • 1School of Geomatics and Urban Spatial Information, Beijing University of Civil Engineering and Architecture, Beijing, China
  • 2National Geomatics Center of China, Beijing, China

Keywords: Web map, Vector tile, Pyramid model, Parallel processing, Apache Sedona

Abstract. With the development of spatial information technology, the amount of geographic information data shows an explosive growth, which puts forward higher demands on the time efficiency and visualization effect of geographic information data release. vector tile maps have become the main map service mode in the fields of Internet maps and GIS industries because of the advantages of its powerful interactive capabilities, efficient data transmission and lower storage costs. In order to construct massive vector tiles quickly and enhance the scalability of vector tiles application, this paper researched the vector tiles construction technology based on the distributed computing framework. Firstly, we introduced the construction process of vector tiles such as pyramid model, data organization and data generalization, Specifically, the data generalization methods of different geometric types were discussed. Second, Apache Sedona, the distributed computing framework were researched and the advantages for vector data processing is introduced. Then, the Sedona based parallel construction technology is proposed, and pipeline optimization of Spark is applied in this process. Third, a comparative experiment to evaluate the performance were conducted, the result showed that the parallel construction technology had obvious performance advantages, the greater the volume and extent of vector data, the greater the advantage.