|Parallel visualization of gigabyte datasets in GeoFEM|
Issei Fujishiro, Yuriko Takeshima, Li Chen, Hiroko Nakamura, Yasuko Suzuki
Parallel visualization of large datasets in GeoFEM is described. Our visualization subsystem supports concurrent visualization with computation, and outputs a simplified small graphic primitive set, which can be displayed by a basic viewer on clients. This subsystem provides many kinds of parallel visualization algorithms for the users to visualize their data from scalar, vector to tensor. Scalar field topology analysis, flow field topology and semi-automatic parameter design are employed to improve the quality of visualization results. We also present a simplification algorithm to reduce the number of output graphic primitives, accounting for both shape attribute and color attribute. The experimental results show the effectiveness of our subsystem.
| PDF (70KB) |