Gpu Accelerated Scalable Parallel Coordinates Plots
1 Chair for Computer Graphics and Visualization, Technical University of Munich, Germany
2 Ludwig Maximilians University Munich
Parallel coordinates are a powerful technique to visually analyze multi-parameter data, i.e., sets of data points with potentially many associated parameter values per data point. When these sets are large, line rendering becomes a severe performance bottle- neck, and since many lines fall into the same pixel the numerical precision of the color bu ff er is quickly reached. We propose a scalable GPU realization of parallel coordi- nates building upon 2D pairwise attribute bins, to significantly reduce the number of lines to be rendered. Our approach comprises a GPU compute pipeline that combines shader-based scattering with atomic increment operations to e ffi ciently count how often a line is drawn. These counts are then used to draw all pairwise sub-plots in the parallel coordinates plot, by analytically calculating the opacity for each count and rendering a line with end points determined by the 2D coordinates of the bin. In this way, frame- bu ff er precision issues that are paramount in classical approaches can be overcome. We demonstrate the e ffi ciency of the proposed realization for visualizing a weather forecast ensemble comprising 2.7 billion data points, each carrying 7 prognostic floating-point variables like temperature, precipitation and pressure, plus spatial and simulation input variables. We compare our pipeline to a rasterization-based approach regarding perfor- mance, and demonstrate interactive brushing at 4 seconds per frame at full HD viewport resolution.