Window Query and Analysis on Massive Spatio-temporal Data

Along with the expansion of computer-based climate simulations, efficient visualization and analysis of massive climate data are becoming more important than ever. In this paper, we try to explore the factors behide climate changes by combining window query and time-varying data mining techniques. With constant query time and acceptable storage cost, the algorithms presented support various queries on 3d time-varying datasets, such as average, min, and max value. A new time-varying data analysis algorithm is given, which is especially suitable for analyzing big data. All these algorithms have been implemented on and integrated into a visual analysis system, with tiled-LCD ultra-resolution display. Experimental results on several datasets from practical applications are presented.