BOUT++: a framework for parallel plasma fluid simulations

Abstract: A new modular code called BOUT++ is presented, which simulates 3D fluid equations in curvilinear coordinates. Although aimed at simulating Edge Localised Modes (ELMs) in tokamak X-point geometry, the code is able to simulate a wide range of fluid models (magnetised and unmagnetised) involving an arbitrary number of scalar and vector fields, in a wide range of geometries. Time evolution is fully implicit, and 3rd-order WENO schemes are implemented. Benchmarks are presented for linear and non-linear problems (the Orszag-Tang vortex) showing good agreement. Performance of the code is tested by scaling with problem size and processor number, showing efficient scaling to thousands of processors. Linear initial-value simulations of ELMs using reduced ideal MHD are presented, and the results compared to the ELITE linear MHD eigenvalue code. The resulting mode-structures and growth-rate are found to be in good agreement (BOUT++ = 0.245, ELITE = 0.239). To our knowledge, this is the first time dissipationless, initial-value simulations of ELMs have been successfully demonstrated.

• The paper introduces BOUT++, a flexible C++ framework for efficient, parallel 3D plasma fluid simulations using finite-difference methods and implicit time evolution.
• BOUT++ supports various fluid models and geometries, demonstrating high accuracy and numerical stability in simulating plasma phenomena through benchmark tests.
• Efficiently scaling on thousands of processors, BOUT++ is a valuable tool for plasma edge dynamics research, including ELMs, and broader fluid dynamics applications.

Overview of "BOUT++: a framework for parallel plasma fluid simulations"

The paper "BOUT++: a framework for parallel plasma fluid simulations" introduces BOUT++, an advanced computational tool designed for simulating fluid equations in plasma environments, specifically aimed at studying the Edge Localised Modes (ELMs) in tokamak devices. This code is constructed to run efficient, parallel plasma simulations in 3D curvilinear coordinates and is adaptable for a variety of fluid models, including both magnetised and unmagnetised plasmas. Its flexibility allows researchers to address numerous fluid dynamic problems across different geometrical configurations.

Key Features and Capabilities

BOUT++ builds upon and extends the original BOUT code, transforming it into a highly modular C++ framework. The code is distinct in its usage of finite-difference methods with fully implicit time evolution, and it incorporates third-order Weighted Essentially Non-Oscillatory (WENO) schemes for handling shocks, enhancing the numerical stability and accuracy in simulating sharp gradients and discontinuities often found in plasma dynamics.

The fundamental components include:

• Time Integration: Utilizes the CVODE solver for implicit integration, which is crucial for dealing with stiff systems without forfeiting time step stability.
• Parallel Processing and Scaling: BOUT++ demonstrates efficient scaling on thousands of processors, showing its potential for large-scale simulations.
• Differential Operations: A wide array of differencing schemes are available, including higher-order central and upwind differencing as well as spectral methods in the periodic dimension.
• Input/Output: Employs Portable Data Binary (PDB) format for data management, which ensures reproducibility and ease of analysis with available C, FORTRAN, and Python bindings.

Numerical Validation and Performance

The paper details several benchmark tests to evaluate the precision and robustness of BOUT++. Notably, the code accurately simulates classical problems such as the resistive drift-wave instability, interchange modes, and the Orszag-Tang vortex problem, each test demonstrating BOUT++'s capability to handle complex plasma phenomena with high numerical fidelity. These results not only validate the mathematical and physical models within BOUT++ but also underscore the framework's applicability for broader fluid dynamics contexts.

Particularly significant is the linear benchmarking of the code against the ELITE linear MHD code for ELM simulations, which demonstrates agreement in growth rates and mode structures, affirming BOUT++'s suitability for cutting-edge plasma research.

Implications and Future Directions

BOUT++ is poised to be a valuable tool in plasma physics research, particularly for those working on understanding and mitigating ELMs in fusion reactors, where such instabilities pose critical challenges for material integrity and plasma containment. The code's versatility suggests potential applications beyond plasma physics, offering a platform for other fluid dynamic simulations that require complex geometrical handling and parallel computation.

Future development of BOUT++ will likely focus on improving the modeling of the vacuum-plasma interface, expanding the types of physical phenomena that can be modeled, and optimizing the parallel computing capabilities further. The ability to scale efficiently with increasing processor numbers opens up possibilities for high-resolution simulations of turbulent plasma and non-linear regime dynamics, crucial for the advancement of futuristic fusion energy solutions.

In summary, BOUT++ presents a comprehensive framework for simulation with a primary focus on plasma edge dynamics. Its design ensures adaptability and efficiency, making it a prominent tool for researchers in computational physics and engineering domains focused on plasma and fluid dynamics.