ABSTRACT High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning. 相似文献
Journal of Geographical Sciences - The border areas of the Tibetan Plateau and the neighboring mountainous areas have a high incidence of earthquakes with a magnitude greater than Ms 5.0, as well... 相似文献
The ecological series of soil animals under the broad-leaved and pine mixed forest in Darlidai Mountainwas studied. Seven sample plots were selected according to different altitude gradients, which belong to different vegeta-tion types. By investigating and analyzing soil animals in every sample plot it is found that there are 45 groups and 1956individuals, which axe involved in 3 phylums, 7 classes, 16 orders, respectively. The altitude is a key factor which af-fects ecological series of soil animals. Both the groups and individuals of soil animals increase with altitude increasingunder certain conditions, which contrastes with ordinary cases, resulting from special micro-climate in studied area. Thegroups and individuls of soil animals are the most under the broad-leaved and pine forest on the top of the mountain, andthe least under Picea-Abies forest in the foot of the mountain. 相似文献
As well known, the methods of remote sensing and Bowen Ratio for retrieving surface flux are based on energy balance closure; however, in most cases, surface energy observed in experiment is lack of closure. There are two main causes for this: one is from the errors of the observation devices and the differences of their observational scale; the other lies in the effect of horizontal advection on the surface flux measurement. Therefore, it is very important to estimate the effects of horizontal advection quantitatively. Based on the local advection theory and the surface experiment, a model has been proposed for correcting the effect of horizontal advection on surface flux measurement, in which the relationship between the fetch of the measurement and pixel size for remote sensed data was considered. By means of numerical simulations, the sensitivities of the main parameters in the model and the scaling problems of horizontal advection were analyzed. At last, by using the observational data acquired in agricultural field with relatively homogeneous surface, the model was validated.