Title :
Optimized Data I/O Strategy of the Algorithm of Parallel Digital Terrain Analysis
Author :
Yan Li ; Wanfeng Dou ; Kun Yang ; Shoushuai Miao
Author_Institution :
Nanjing Normal Univ., Nanjing, China
Abstract :
In order to improve the analysis performance of the algorithm of parallel digital terrain analysis, a series of parallel tasks should be proceed by the appropriate partition and scheduling strategy, but the algorithm of data I/O strategy can also be optimized. Based on the conventional data I/O model, the DEM data I/O strategy of the parallel algorithm for digital terrain can be optimized by the parallel pipeline method. Then it can further enhance the efficiency of the parallel algorithm of digital terrain analysis. Take the maximum slope algorithm for example, it was conducted many experiments in the small-scale cluster environment. The results show that the parallel algorithm based on the optimized data I/O strategy can improve the analysis efficiency of the previously conventional algorithm.
Keywords :
data models; geographic information systems; parallel algorithms; pipeline processing; terrain mapping; DEM data I/O strategy; data I/O model; maximum slope algorithm; optimized data I/O strategy; parallel algorithm; parallel digital terrain analysis; parallel pipeline method; parallel tasks; partition strategy; scheduling strategy; small-scale cluster environment; Algorithm design and analysis; Clustering algorithms; Parallel algorithms; Partitioning algorithms; Pipelines; Writing; the asynchronous data I/O; the digital terrain analysis; the parallel computation; the parallel pipeline;
Conference_Titel :
Distributed Computing and Applications to Business, Engineering and Science (DCABES), 2014 13th International Symposium on
Conference_Location :
Xian Ning
Print_ISBN :
978-1-4799-4170-4
DOI :
10.1109/DCABES.2014.10
