作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (22): 83-85. doi: 10.3969/j.issn.1000-3428.2007.22.029

• 软件技术与数据库 • 上一篇    下一篇

基于MapReduce的分布式光线跟踪的设计与实现

郑欣杰,朱程荣,熊齐邦   

  1. (同济大学计算机科学与技术系,上海 200331)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-20 发布日期:2007-11-20

Design and Implementation of Distributed Ray Tracing Using MapReduce

ZHENG Xin-jie, ZHU Cheng-rong, XIONG Qi-bang   

  1. (Dept. of Computer Science and Technology, Tongji University, Shanghai 200331)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-20 Published:2007-11-20

摘要: 提出了基于MapReduce架构实现分布式光线跟踪渲染的方案。该方案基于Hadoop实现,利用MapReduce架构简化了分布式程序设计。使用分布式计算进行光线跟踪,充分利用了现有低端硬件设备的处理能力。实验表明,该方案通过并行计算大大加快了渲染速度。

关键词: MapReduce, Hadoop, 分布式计算, 光线跟踪

Abstract: An approach based on Hadoop is presented to speed up ray tracing by utilizing MapReduce. It improves the rendering speed considerably by employing distributed computing, and simplifies the design of the distributed application by using MapReduce paradigm. Hadoop implementation provides a convenient and flexible framework for distributed computing on a cluster of commodity machines. The design and implementation is presented. The results and some related discussions are also given.

Key words: MapReduce, Hadoop, distributed computing, ray tracing

中图分类号: