Abstract:Scientists who want to use micromagnetic simulation packages, need to learn how to express the micromagnetic problem of interest in a “language” such as a configuration file, script or GUI-clicks that the simulation software understands. This language varies from software to software. If the researchers need to use another package, they need to learn a new language to re-express their (unchanged) micromagnetic problem for the next software. For research-project specific pre- or post-processing, scientists often need to write additional software. We propose the vision of a unified micromagnetic modeling (UMM) approach with which researchers can express the micromagnetic problem once and from which multiple simulation packages can be instructed automatically to carry out the actual numerical problem solving. Furthermore, by providing defined interfaces that are embedded in the Python data science ecosystem and which are used to communicate with the simulation packages, we can create an open research framework in which simulation runs and additional computation can be arbitrarily combined and orchestrated. Where analysis tools are missing from simulators, these can conveniently be created at Python level. Advantages of this approach include reduced effort for scientists to familiarize themselves with multiple simulation configuration languages, easier exploitation of complementary features of the different simulation packages, the ability to compare results computed with different simulation packages more easily, and the option to easily extend analysis functionality of the existing simulators. With recent updates of Ubermag we present a prototype of such a UMM framework. Ubermag provides a unified interface (expressed in Python) to solve micromagnetic problems using, currently, OOMMF and mumax3. After a simulation has finished, the results are made available to the researcher for analysis within the Python ecosystem of scientific and data science libraries. We discuss the current state of capabilities and challenges associated with the proposed approach.

magnum.np: a PyTorch based GPU enhanced finite difference micromagnetic simulation framework for high level development and inverse design

mumax+: extensible GPU-accelerated micromagnetics and beyond

Finite Difference Micromagnetic Solvers with the Object Oriented MicroMagnetic Framework (OOMMF) on Graphics Processing Units

Micromagnetics on high-performance workstation and mobile computational platforms

Parallelized Implementation of the Finite Particle Method for Explicit Dynamics in GPU

MagTense: a micromagnetic framework using the analytical demagnetization tensor

A full-fledged micromagnetic code in less than 70 lines of NumPy

Ubermag: Towards more effective micromagnetic workflows

Grace: A cross-platform micromagnetic simulator on graphics processing units

NeuralMAG: Fast and Generalizable Micromagnetic Simulation with Deep Neural Nets

GPU-accelerated micromagnetic simulations using cloud computing

The design and verification of MuMax3

Micromagnetic Modeling Simulations And Applications

Accelerating micromagnetic and atomistic simulations using multiple GPUs

Parallel-in-Time Integration of the Landau-Lifshitz-Gilbert Equation with the Parallel Full Approximation Scheme in Space and Time

TorchMD: A deep learning framework for molecular simulations

Vision for unified micromagnetic modeling (UMM) with Ubermag

Virtual Micromagnetics: A Framework for Accessible and Reproducible Micromagnetic Simulation

User interfaces for computational science: a domain specific language for OOMMF embedded in Python

Performance Portable Monte Carlo Neutron Transport in MCDC via Numba