Interactive Analytics of Massive Spatial Vector Data via Display-Driven Computing

Abstract

In the big data era, real-time analysis of large-scale spatial data remains a challenging problem since the computational scales of conventional data-driven methods expand rapidly with data volumes. This paper presents display-driven computing (DisDC), a new computing model for interactive analysis of large-scale spatial vector data. The computing units in DisDC are screen pixels for display rather than spatial objects in traditional data-driven methods, and the core task of DisDC is to judge the spatial relationships between pixels and spatial objects. As the number of pixels for display is stable, DisDC is less sensitive to data volumes and can be used to provide real-time data analysis of large-scale spatial vector data. Experiments show that DisDC is capable of handling billion-scale spatial vector data, and an online demonstration is provided (https://github.com/MemoryMmy/DisDC).