Abstract: A memory model suitable for dedicated architecture of volume rendering based on Ray Casting algorithm is proposed. Volume data is divided into subcubes based on the number of processors. A subcube is distributed to the corresponding processing unit according to the space coordinates. Each voxel of a subcube is stored in a given memory unit. The way of the assignment of subcube to different processors is introduced and the addressing scheme of the voxels is described.

Key words: Volume rendering, Architecture, Memory model, Ray casting

摘要： 提出了一种适合Ray Casting算法的体绘制专用体系结构的存储模型。根据处理器数目的不同，体数据被划分为不同的子体。子体依据其空间坐标位置的不同被分配到不同处理单元，子体内的体素被分配到相应处理单元的存储器的对应位置。说明了子体在处理器间的分配方式以及体素在存储器内的编址和寻址方式。

关键词: 体绘制, 体系结构, 存储模型, 光线投射