The extraction of feature points from DEM geographic data in Cloud Computing environment

As the technology of obtaining geographic data continues to update, we can get geographic data more conveniently and efficiently. However, handling a large number of geographic data becomes the bottleneck of geography. As a new parallel processing technology, Cloud Computing shows an excellent performance in the big data calculation and network storage. Digital Elevation Model (DEM) shows the fluctuation of the terrain feature and includes the structure information of landform, for example the valley points and the peak points. The points play a very important position when we reconstruct the surface and reappear the surface under the multi-scale. In order to solve the problem of extracting the feature points´ inefficiency in a large-scale regular grid, the paper proposed one thinking which is that we create one virtual net on the Windows azure and use the Message Passing Interface (MPI) to construct a parallel environment to extract the feature points in the regular grid which is incised with one design. By comparing the time parallel computing and serial computing consume, we can get that the parallel computing can improve the efficiency. This thinking can help us apply the Cloud Computing on the analysis of large-scale geography data.