DDDDRRaW: A prototype toolkit for distributed real-time rendering on commodity clusters
Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001
We describe DDDDRRaW, a prototype toolkit for real-time rendering on clusters of commodity computers. In contrast to most work on cluster computing, DDDDRRaW supports a repeated, low-latency computation, the drawing of frames, which must take place on a time scale of 30-100 ms. DDDDRRaW employs Image Layer Decomposition, a rendering-specific work partitioning algorithm described and evaluated via simulation in  . In this paper, we address implementation issues not confronted in the design
... ted in the design of ILD. In particular, one important issue that must be confronted is how to exploit the potential parallelism afforded by the multiple hardware resources of each node: the CPU, the network adapter, and the video card. We evaluate DDDDRRaW's live performance on two small workstation clusters representing different points in the technology spectrum. Our results show that DDDDRRaW effectively exploits cluster resources to improve real-time rendering performance and should scale well to moderately sized clusters. While the commodity cluster is an attractive platform because of its ubiquity, low cost, and ease of expandability, it presents a number of challenges. In particular, a cluster-based distributed real-time renderer must be structured to: (i) use the multiple hardware graphics accelerators in the cluster to increase rendering performance over what is achievable by a sequential renderer that makes use of an accelerator, (ii) only impose overheads compatible with the 30-100 ms perframe compute load of the application, (iii) minimize variance in the frame rate because variance is important to a user's perceived quality of service, and (iv) decouple communication bandwidth requirements from the complexity of the scene and the number of nodes used to achieve application and system scalability. £ DDDDRRaW is pronounced draw and stands for Distributed 3D Real-Time Rendering at Washington.