| 线程级并行计算在图形渲染引擎中的研究 |
| |
| 引用本文： | 赵建斌,李灵巧,杨辉华.线程级并行计算在图形渲染引擎中的研究[J].计算机工程与设计,2011,32(12):4143-4146. |
| |
| 作者姓名： | 赵建斌 李灵巧 杨辉华 |
| |
| 作者单位： | 桂林电子科技大学,计算机科学与工程学院,广西桂林541004 |
| |
| 摘 要： | 针对并行计算技术在当前图形渲染系统中应用层面较浅的问题,为提高多核平台下图形应用程序CPU利用率,提出了一种新的Fork-Join多线程渲染方案。对当前流行的开源图形渲染引擎——OGRE引擎的渲染框架进行了多线程优化,用OpenMP方法对图形引擎的逻辑帧实现并行化,根据Win32线程库和DirectX11的多线程支持提出了一种渲染帧并行化方法,并将其应用于图形引擎。在多核平台上的实验结果表明,该方案能有效提高渲染速率和CPU利用率,改善CPU负载均衡。
|
| 关 键 词： | 并行计算 多核平台 多线程 图形渲染 延迟执行 |
Research on graphic rendering engine with thread-level parallelism method |
| |
| ZHAO Jian-bin,LI Ling-qiao,YANG Hui-hua.Research on graphic rendering engine with thread-level parallelism method[J].Computer Engineering and Design,2011,32(12):4143-4146. |
| |
| Authors: | ZHAO Jian-bin LI Ling-qiao YANG Hui-hua |
| |
| Affiliation: | ZHAO Jian-bin,LI Ling-qiao,YANG Hui-hua(Institute of Computer Science and Engineering,Guilin University of Electronic Technology,Guilin 541004,China) |
| |
| Abstract: | Aimed at the problems of the low level application of parallel computing technology in graphic rendering systems and to increase the CPU utilization of graphic application program in multi-core systems,a new fork-join multithreaded rendering method is presented.A multithreaded optimization is applied to a popular open sourced graphic rendering engine named OGRE.Firstly,the logical frame is parallelized by the method of OpenMP.Then,according to the Win32 library and the DirectX11 multithread support,a new me... |
| |
| Keywords: | parallel computing multi-core platform multithreading graphics rendering deferred context |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
