作者: Jeong-Gyu Kim , Woong-Tae Kim , Eve C. Ostriker , M. Aaron Skinner
关键词: Regular grid 、 Computational science 、 Scaling 、 Radiation pressure 、 Parallel algorithm 、 Radiation 、 Radiative transfer 、 Ray tracing (graphics) 、 Star formation 、 Physics 、 Optics
摘要: We present an implementation of adaptive ray tracing (ART) module in the Athena hydrodynamics code that accurately and efficiently handles radiative transfer involving multiple point sources on a three-dimensional Cartesian grid. adopt recently proposed parallel algorithm uses non-blocking, asynchronous MPI communications to accelerate transport rays across computational domain. validate our through several standard test problems including propagation radiation vacuum expansions various types HII regions. Additionally, scaling tests show cost full trace per source remains comparable update up $\sim 10^3$ processors. To demonstrate application ART implementation, we perform simulation star cluster formation marginally bound, turbulent cloud, finding its efficiency is $12\%$ when both pressure forces photoionization by UV are treated. directly compare computed from scheme with M1 closure relation. Although schemes yield similar results large scales, latter unable resolve field near individual sources.