飞行器大数据量CAD模型并行绘制

系统仿真学报 ›› 2016, Vol. 28 ›› Issue (9): 2049-2054.

飞行器大数据量CAD模型并行绘制

谈敦铭¹, 曹国廷¹, 郎娟芳¹, 杨朔²

1.空军第一航空学院, 河南信阳 464000;
2.中国人民解放军95903部队,湖北武汉 430331

收稿日期:2015-05-18 修回日期:2015-07-24 出版日期:2016-09-08 发布日期:2020-08-14
作者简介:谈敦铭(1983-), 男, 甘肃兰州, 博士, 讲师, 研究方向为虚拟现实,图形学;曹国廷(1972-), 男, 山东烟台, 本科,副教授, 研究方向为虚拟仿真、发动机原理与构造。
基金资助:
教育部新世纪优秀人才支持计划(NCET- 10-0036)

Parallel Rendering of Massive Aircraft CAD models

Tan Dunming¹, Cao Guoting¹, Lang Juanfang¹, Yang Shuo²

1. The First Aeronautical Institute of PLA Air Force, Xinyang, 464000, China;
2. No. 95903 Unit of PLA, Wuhan, 430331, China

Received:2015-05-18 Revised:2015-07-24 Online:2016-09-08 Published:2020-08-14

摘要/Abstract

摘要： 针对飞行器CAD模型的数据量非常大,难以实现高分辨率实时绘制的问题,提出了一种基于拼接屏的多CPU并行绘制方法。集群绘制节点之间通过基于Sort-first的拼接屏并行绘制,将CAD模型构建动态BVH树,快速剔除单个节点视锥体外不可见的部分以降低负载,绘制结果通过拼接屏输出显示。集群绘制节点内部利用多核处理器开展数据调度、可见性剔除、BVH树更新、绘制等多线程并行进一步提高性能。实验结果表明,利用15个绘制节点并行绘制包含1 700万三角面片的飞行器大数据量CAD模型,可实时绘制分辨率高达6800*2304像素的图像。

关键词: 大数据量CAD模型, 并行绘制, 机群, 多线程

Abstract: A parallel rendering method was presented with PC clusters in order to render massive aircraft CAD models in real time with high resolution. PC clusters were used for sort-first based rendering, BVH tree was built for efficient frustum culling, DID screens were utilized for high resolution display. In addition, multi-CPUs were exploited for thread based parallel out-of-core paging, frustum culling, BVH tree building and rendering. The experimental results show that by this parallel rendering PC clusters consists of 15 PC nodes, aircraft CAD models with 1.7 million of triangles can be rendered in real-time with resolution of 6800* 2304 pixels.

Key words: massive CAD models, parallel rendering, cluster, multi-threads

中图分类号:

TP391.7

谈敦铭, 曹国廷, 郎娟芳, 杨朔. 飞行器大数据量CAD模型并行绘制[J]. 系统仿真学报, 2016, 28(9): 2049-2054.

Tan Dunming, Cao Guoting, Lang Juanfang, Yang Shuo. Parallel Rendering of Massive Aircraft CAD models[J]. Journal of System Simulation, 2016, 28(9): 2049-2054.

参考文献 11

[1]	Stephens A, Boulos S, Bigler J, et al.An application of scalable massive model interaction using shared memory systems [C]// Proceedings of the Eurographics Symposium on Parallel Graphics and Visualization. Aire-la-Ville, Switzerland: Eurographics Association Press, 2006: 19-26.
[2]	Tian Fenglin, Hua Wei, Dong Zilong, et al.Adaptive voxels: interactive rendering of massive 3D models[J]. The Visual Computer(S0178-2789), 2010, 26(6): 409-419.
[3]	Peng C and Cao Y. GPU-based streaming for parallel level of detail on massive model rendering [D]. Virginia, USA: Virginia Tech. Computer Science, 2011.
[4]	Humphreys G, Houston M, Ng R, et al.Chromium: a stream-processing framework for interactive rendering on clusters[J]. ACM Transactions on Graphics (TOG)(S0730-0301), 2002, 21(3): 693-702.
[5]	Eilemann S, Makhinya M, Pajarola R.Equalizer: A Scalable Parallel Rendering Framework[J]. IEEE Transactions on Visualization and Computer Graphics:(S1077-2626), 2009, 15(3): 436-452.
[6]	Peng H, Xiong H, Shi J.Parallel-SG: research of parallel graphics rendering system on PC-Cluster [C]// Proceedings of the 2006 ACM International Conference on Virtual Reality Continuum and Its Applications. New York, USA: ACM Press, 2006: 27-33
[7]	殷萍. Terrain-VDR: 基于Out-of-Core的交互式地形并行绘制框架 [D]. 杭州: 浙江大学, 2006.
[8]	陈学亮. 基于多核平台优化的OGRE 3D渲染引擎 [D]. 杭州: 浙江大学, 2007.
[9]	王攀, 刘华海, 李思昆, 等. 无合成并行绘制算法中的自适应屏幕分割策略[J]. 系统仿真学报, 2012, 24(1): 99-103.
[10]	刘华海, 王攀, 蔡勋, 等. 面向多核CPU多GPU的节点内并行混合绘制模型[J]. 系统仿真学报, 2012, 24(1): 94-98.
[11]	谈敦铭, 赵罡. 面向装配的飞行器超大模型实时可视化技术[J]. 计算机辅助设计与图形学报, 2012, 24(5): 590-597.