Multi-paradigm Geocomputing Techniques

Different geocomputing models are available for different data types. Of course, juggling multiple applications will increase the technical complexity of the system and make performance optimization more difficult. High-performance GIS is designed to solve this problem. By combining the advantages of computing architectures such as message passing, mapping-reduction, in-memory computing, real-time computing, and stream computing, a multi-paradigm high-performance geocomputing solution is realized. This chapter will introduce the current development status of the multi-paradigm geocomputing technology, which use parallel computing technology to improve the efficiency of geocomputing algorithms. These technologies can be further catagorized into three subdivisions: parallel geocomputing based on message passing interface, parallel geocomputing based on distributed file system, and parallel geocomputing based on distributed memory.