RAMP: A prototype for massively parallelized Monte Carlo raytracing

Abstract

Simulated optimization of neutron instrumentation via Monte Carlo raytracing (MCR) has played a vital role in instrument design in recent decades. With the increasing complexity of modern instruments, however, optimization is becoming more computationally demanding. Graphics processing unit (GPU) acceleration is a promising strategy to reduce the hardware cost and execution time of simulations, thanks to the massively parallelizable nature of the MCR algorithm. In this proceedings, we present RAMP, a prototype MCR package employing GPU acceleration through the OpenCL application program interface (API). Our preliminary results demonstrate the potential to run simulations on a typical workstation which previously would have required access to a cluster.