摘要
根据图形处理器的最新可编程单元Vertex Shader和Pixel Shader的体系结构和单机Direct3D9应用程序的执行流程,提出支持Shader的Direct3D9应用程序在图形集群的透明并行化策略.图形集群的节点划分为资源分配和资源绘制节点,资源分配节点通过截取绘制接口将应用程序实时转换为6类绘制资源,包括命令流、Vertex Shader、Pixel Shader、顶点流、索引流和纹理流.资源绘制节点根据绘制资源的描述信息和资源数据重构出Direct3D9的绘制命令.图形集群中的所有绘制节点都保留全部的绘制资源,并且通过计算基于多流模式场景数据在屏幕空间的包围盒进行绘制任务划分.实验证明,使用,这种策略完全可以实现支持Shader的Direct3D9应用程序透明并行化.相对于单机绘制,基于图形集群的并行图形绘制不仅提高绘制性能而且得到较高绘制加速比.
Vertex shader and pixel shader are the newest programmable units of graphics processing unit.According to their architecture and Direct3D9 application execution flow on single machine,transparent shader-based Direct3D9 application running on graphics cluster parallelization strategy is proposed.Graphics cluster can be divided into rendering resource distributing node and rendering resource rendering node.Among them,distributing node is responsible for converting Direct3D9 application to six kinds of rendering resource,including command stream,vertex shader,pixel shader,vertex stream,index stream and texture stream.Rendering node is responsible for reconstructing Direct3D9 rendering command based on the description information and resource data of rendering resource.All of the rendering nodes reserve the entire rendering resource and distribute rendering task through computing the bounding box of multi-stream based scene data in the screen space.Experimental results show that this strategy can realize transparent shader-based Direct3D9 application parallelization.In contrast to single machine rendering,parallel graphics rendering based on graphics cluster can not only promote rendering performance but also achieve higher rendering speedup.
出处
《计算机研究与发展》
EI
CSCD
北大核心
2007年第10期1673-1681,共9页
Journal of Computer Research and Development
基金
国家"九七三"重点基础研究发展规划基金项目(2002CB312105)
国家自然科学基金重点项目(60533080)
