I/O Optimized MultiResolution Data Organization

摘要：

Large scale simulation data challenges the ability of visualization. It becomes an important research content that how to reorganize the large scale data into multiresolution representation and improve the interactive performance of visualization. Considering the data block size produced by large scale simulation mismatches the need of visualization, A novel parallel algorithm of I/O optimized multiresolution data organization is introduced, and it reorganizes simulation data with lower memory need into multiresolution form which matching the optimized I/O size. This technique has been tested on large scale simulation data, its output performance is better than directly parallel write, and interactive visualization performance improves significantly.

作者: Weichao Shen Pingjun Xu Fang Xia

作者单位: High Performance Computing Center, Institute of Applied Physics and Computational Mathematics, Beijing, 100088, China

会议类型: 国际会议

会议名称: 第十届中国虚拟现实年会

会议地点: 上海

会议语种:英文

页码: 648-653

在线出版日期: 2010-10-20（万方平台首次上网日期，不代表论文的发表时间）