MC/DC: Monte Carlo Dynamic Code

MC/DC is an open-source, Python-based Monte Carlo radiation transport software package that combines rapid methods development with scalable execution on modern high-performance computing systems. It supports execution across CPUs and GPUs while providing a flexible environment for developing and testing new transport algorithms.

MC/DC is intended for researchers developing new Monte Carlo transport methods, including variance reduction techniques, sensitivity and uncertainty quantification methods, as well as high-performance computing algorithms. It also provides an accessible platform for students learning Monte Carlo radiation transport methods and modern code development.

MC/DC supports continuous-energy and multi-group neutron transport calculations, including fixed-source and eigenvalue simulations on constructive solid geometry (CSG) models. For continuous-energy transport, MC/DC converts ACE-format nuclear data libraries into its native HDF5 format. Photon, electron, proton, and other charged-particle transport capabilities are currently under development as part of the ongoing expansion of MC/DC into a comprehensive multi-particle radiation transport software package.

MC/DC’s Python interface enables rapid prototyping and iterative development, while its Numba-based compilation framework delivers high performance without sacrificing portability. Harmonize provides a GPU execution framework, and MPI4Py enables distributed-memory parallelism across large HPC systems. In addition to desktop and workstation systems, MC/DC has been demonstrated on large heterogeneous supercomputers, including Lassen (IBM POWER9 and NVIDIA Volta V100) and Tuolumne (AMD MI300A APU).

MC/DC was initiated by the Center for Exascale Monte Carlo Neutron Transport (CEMeNT), a Focused Investigatory Center of the Predictive Science Academic Alliance Program–III (PSAAP-III). Development is now led by the Center for Advancing the Radiation Resilience of Electronics (CARRE), a Predictive Simulation Center of PSAAP-IV.

MC/DC is released under the BSD 3-Clause license and welcomes community contributions through GitHub.

Recommended citation

Morgan, Joanna Piper, et al. “Monte Carlo/Dynamic Code (MC/DC): An accelerated Python package for fully transient neutron transport and rapid methods development.” Journal of Open Source Software 9.96 (2024): 6415. https://joss.theoj.org/papers/10.21105/joss.06415

Contents

User Documentation

• Installation Guide
• User’s Guide
• Input Definition
• Example Problems

Developer Documentation

• Contribution Guide
• Theory Guide

References

• Publications