A Scalable parallel ray tracing algorithm for satellite Imagery

摘要：

Ray-tracing, can produce high-quality images, however, the use of ray-tracing has been limited due to its high demands on computational power and memory bandwidth, especially in the case of satellite imagery. In this paper, we propose a scalable parallel ray tracing algorithm of satellite imagery on a cluster of multi-core architecture. The algorithm combines demand-driven and data-driven models in order to reduce communication on the cost of maintaining redundant data. In order to make the most of computers parallelism and increase computational efficiency, it combines two levels of parallelisms, the TLP parallelism brought by the many core architecture and the inter-node parallelism via MPI and OpenMP. Experiment results show that the algorithm is highly scalability.

关键词： Digital elevation model (DEM) Ray tracing parallel OpenMP MPI

作者: Li Yu-Gang Zhao Kun Huang Yu-Qing Qiu Zhen-Ge Liu Zhi-Yong

作者单位: Beijing Lab of Intelligent Information Technology,School of Computer Science,BIT Key Laboratory of C Institute of Computing Technology,Chinese Academy Sciences Beijing,China

会议类型: 国际会议

会议名称: 2011 3rd IEEE International Conference on Computer Research and Development(ICCRD 2011)(2011第三届计算机研究与发展国际会议)

会议地点: 上海

会议语种:英文

页码: 151-155

在线出版日期: 2011-03-11（万方平台首次上网日期，不代表论文的发表时间）