Abstract

Charles B. Cameron, Parallel Ray Tracing Using the Message Passing Interface (MPI), submitted for publication to IEEE Transactions on Instrumentation and Measurement on 19 September 2006.  Accepted on 5 September 2007 for publication with revisions. 

Ray tracing software is available both for lens design and for general optical systems modeling. It tends to be designed to run on a single processor and can be very time-consuming if the number of rays traced is large. Previously, multiple digital signal processors (DSP) have been used to perform such simulations. This approach is attractive because DSPs are inexpensive and the time saved through parallel processing can be significant. In this paper we report a nearly linear relationship between the number n of processors and the rate of ray tracing with as many as 839 processors operating in parallel on the Naval Research Laboratory's Cray XD-1 computer with the Message Passing Interface (MPI). In going from one processor to 839 we achieved an efficiency of 97.9 % and a normalized ray-tracing rate of 6.95 · 106 rays · surfaces /(s · processor) in a system with 22 planar surfaces, two paraboloid reflectors, and one hyperboloid refractor.  The need for load-balancing software was obviated by the use of a prime number of processors.


Return to CDR Cameron's Home Page
Send E-mail