The Era of Big Spatial Data: A Survey

Abstract

The recent explosion in the amount of spatial data calls for specializedsystems to handle big spatial data. In this survey, we summarizethe state-of-the-art work in the area of big spatial data. We categorizethe existing work in this area according to six different angles, namely,approach, architecture, language, indexing, querying, and visualization.1 The approaches used to implement spatial query processing can becategorized as on-top, from-scratch and built-in approaches. 2 Theexisting works follow different architectures based on the underlyingsystem they extend such as MapReduce, key-value stores, or parallelDBMS. 3 The high-level language of the system is the main interfacethat hides the complexity of the system and makes it usable fornon-technical users. 4 The spatial indexing is the key feature of manysystems which allows them to achieve orders of magnitude performancespeedup by carefully laying out data in the distributed storage. 5 Thequery processing is at the heart of all the surveyed systems as it definesthe types of queries supported by the system and how efficiently theyare implemented. 6 The visualization of big spatial data is how thesystem is capable of generating images that describe terabytes of datato help users explore them. This survey describes each of these components,in detail, and gives examples of how they are implemented inexisting systems. At the end, we give case studies of real applicationsthat make use of these systems to provide services for end users.