 
Summary: Scalable Monte Carlo Image Synthesis
Alan Heirich a;b and James Arvo b
a Center for Advanced Computing Research
b Department of Computer Science
California Institute of Technology, Pasadena, CA 91125
This paper describes a scalable photorealistic renderer that is de
signed to render scenes of arbitrary complexity on computer sys
tems of arbitrary size. The rendering algorithm is a Monte Carlo
method to compute approximate solutions of the rendering equa
tion. The software implementation uses a diffusive load balancing
method coupled with a message driven concurrent pipeline. Mea
sured performance in rendering replicated models on up to 256
computers shows scaling efficiencies as high as 99 percent. Simple
extensions will partition extremely large models across physically
distributed memory as well as perform outofcore calculations.
1 Introduction
In recent years scalable parallel processors (SPPs) have become readily avail
able to solve computationally intensive problems in science and engineering.
Experience with these applications has shown that for certain classes of prob
lems these SPPs routinely achieve speedups close to a factor of P using P
