Load Balancing for Multi-Projector Rendering Systems

Multi-projector systems are increasingly being used to provide large-scale and high-resolution displays for next-generation inter-active 3D graphics applications, including large-scale data visual-ization, immersive virtual environments, and collaborative design. These systems must include a very high-performance and scalable 3D rendering subsystem in order to generate high-resolution images at real time frame rates. This paper describes a sort-first based par-allel rendering system for a scalable display wall system built with a network of PCs, graphics accelerators, and portable projectors. The main challenge is to develop scalable algorithms to partition and assign rendering tasks effectively under the performance and func-tionality constrains of system area networks, PCs, and commodity 3-D graphics accelerators. We have developed three coarse-grained partitioning algorithms and incorporated them into a working pro-totype system. This paper describes these algorithms and reports our initial experimental results aimed at investigating the feasibility of constructing a sort-first rendering system using a network of PCs and evaluating algorithmic trade-offs and performance bottlenecks within such a system. Our results indicate that the coarse-grained characteristics of the sort-first architecture are well suited for con-structing a parallel rendering system running on a PC cluster. Key Words: Immersive display systems, cluster computing, paral-lel rendering, load balancing