An index model of global subdivision in cloud computing environment

Cloud computing has been widely used in various fields. A typical Google-like infrastructure of cloud computing is usually supported by the following three technologies: a scalable distributed file system like GFS, a structured key-value data store mechanism like Bigtable, and a parallel programming model like MapReduce. This paper presents a multi-level index model based on global subdivision grid in cloud computing environment. Both remote sensing images and spatial objects are indexed. It discusses the design, structure, retrieval and comparison of the index model. The spatial retrieval contains four types: point retrieval, range retrieval, topological retrieval and special grid retrieval. This paper also introduces the design of a prototype system implementing the proposed model, which consists of storage and server layer, service layer, management layer, and application layer.