Abstract—The underground spatial-temporal data of an urban is so large that it is impossible to process them by PC. A PC cluster or a supercomputer is more suitable for them. In order to fully make use of multiple computing nodes, the parallel programs are needed. Traditional parallel programs are usually developed in parallel computing mode such as using MPI which is more complex in programming than hadoop a cloud computing mode. Underground spatial-temporal data has four dimensions including temporal dimension and spatial three dimensions. The relationship among different dimensions is little, which is a very key feature suitable for hadoop to process. In a word, vast underground spatial-temporal data is difficult to be deal with serial computing, but is feasible by parallel computing or cloud computing, and underground spatial-temporal data cloud computing is easier in programming and more scalable and more tolerant but as fast as underground spatial-temporal data parallel computing.

Index Terms—spatial-temporal data, cloud computing, underground

Cite: Renfeng A. Xu and Dingju B. Zhu , "An Approach for Processing Underground Spatial-Temporal Data by Cloud Computing ," Journal of Automation and Control Engineering, Vol.1, No.2, pp. 164-165, June 2013. doi: 10.12720/joace.1.2.164-165