AI summary · arXiv: 2009.02171v2 · physics.comp-ph

Parallel finite volume simulation of the spherical shell dynamo with pseudo-vacuum magnetic boundary conditions

Liang Yin, Chao Yang, Shi-Zhuang Ma, Ying Cai, Keke Zhang
Original paper · DOI 10.1016/j.jcp.2021.110866 Submitted to arXiv on 04 Sep. 2020 Published in Journal of Computational Physics (February 2022) License: NONEXCLUSIVE-DISTRIB 1.0

How to read this summary? Illustrated guide

The essentials

The paper presents a parallel finite-volume spherical-shell dynamo solver that reproduces two pseudo-vacuum benchmarks and scales to 10,368 cores, with a two-level preconditioner improving pressure-system performance.

Question

Can a local finite-volume method efficiently and accurately simulate convection-driven dynamos in a rapidly rotating spherical shell with pseudo-vacuum magnetic boundaries? The authors also investigate whether a multilevel preconditioner improves the parallel scalability of the linear solvers.

Method

The authors discretize the MHD dynamo equations on a collocated, quasi-uniform cubed-sphere grid using a second-order finite-volume scheme and second-order approximate-factorization time integration; a pseudo-pressure enforces the magnetic divergence constraint. They solve the resulting two linear systems with preconditioned GMRES and test two benchmark cases at four resolutions, then assess one- and two-level restricted additive Schwarz preconditioners on Sunway TaihuLight using up to 10,368 processor cores.

Findings

  • Result The spatial discretization and time integration achieve second-order accuracy in the reported convergence tests.
    See the passage in the paper
    It can be seen that the spatial discretization is second-order accurate.
  • Result For benchmark P5, the computed quantities agree closely with the recommended results, with discrepancies below 1% on grids G64, G80, and G96 except for drift frequency.
    See the passage in the paper
    From Table 2, it is seen that the discrepancy is less than 1% for all quantities on grid G64, G80 and G96 except the drift frequency $\omega$.
  • Result For benchmark P8, the computed quantities agree with the pseudospectral benchmark, and relative errors decrease as resolution increases.
    See the passage in the paper
    It can be seen from Table 3 that our results are in good agreement with the benchmark pseudospectral solution and the accuracy is comparable to the existing finite volume results.
  • Result The two-level preconditioner reduces PPBE iterations and compute time relative to the one-level method in the strong-scaling tests.
    See the passage in the paper
    In terms of the compute time of PPBE, the two-level preconditioner is about 20%–47% faster than the one-level, which indicates a noticeable improvement of computational efficiency.
  • Authors’ interpretation The authors report good scalability for VTBE and improved PPBE weak scalability with the two-level preconditioner, while identifying the coarse-grid solve as the main performance bottleneck.
    See the passage in the paper
    For the solver of PPBE, a noticeable improvement in the computational efficiency and the weak scalability by the two-level preconditioner is observed, comparing to the one-level method.

Limitations

  • The study uses pseudo-vacuum magnetic boundaries rather than the insulating condition, and the authors identify implementing the insulating boundary and inner-core singularity as future work.
    See the passage in the paper
    To extend our code to the full dynamo problem, the implementations of the insulating boundary condition and the singularity in the inner core should be taken into consideration in the future.
  • The two-level method's scaling is limited in part by the coarse-grid linear solve, which the authors identify as the dominant and non-scalable component.
    See the passage in the paper
    Furthermore, the major time-consuming and non-scalable part lies with the linear system solution on the coarse mesh.

Why it matters

The authors argue that their cubed-sphere finite-volume solver provides accurate local-method solutions for two pseudo-vacuum dynamo benchmarks while supporting large parallel runs. The multilevel restricted additive Schwarz approach improves the efficiency and weak scalability of the pressure-related system, helping extend local dynamo simulation to larger computations.

Written by AI from the paper. Passages are checked word for word against the text we analysed; the rest may contain errors.

AI-generated keywords: magnetohydrodynamic dynamo; finite-volume method; pseudo-vacuum boundary conditions; parallel scalability; restricted additive Schwarz preconditioner

Read further

  • Comprehensive Summary
  • Layman's Summary
  • Blog article
  • Original abstract
The paper develops and evaluates a parallel finite-volume solver for convection-driven magnetohydrodynamic dynamo simulations in a rapidly rotating spherical shell. It investigates whether a local discretization can produce accurate benchmark solutions, scale to many processor cores, and maintain a divergence-free magnetic field. The solver uses pseudo-vacuum magnetic boundary conditions, which set the tangential magnetic field to zero at the shell boundaries. These conditions are simpler and less costly for local methods than insulating boundaries, but they do not represent the same physical boundary condition. The spatial discretization is a second-order finite-volume scheme on a collocated, quasi-uniform cubed-sphere grid. A pseudo-pressure term is added to the magnetic induction equation so that a projection step can correct the intermediate magnetic field and enforce its solenoidal constraint. Time integration uses a second-order approximate-factorization scheme. Nonlinear terms are treated explicitly with a second-order Adams–Bashforth formula, while linear terms are treated with Crank–Nicolson. Each time step generates two linear systems: one for velocity, temperature, and magnetic field, and another for pressure and magnetic pseudo-pressure. Both systems are solved using preconditioned GMRES. To improve parallel performance, the authors develop a restricted additive Schwarz preconditioner that combines local subdomain solves with a coarse-grid correction. Convergence tests support the scheme’s formal accuracy, with both spatial and temporal errors exhibiting second-order convergence. The authors also evaluate two established pseudo-vacuum dynamo benchmarks. For case P5, the computed results approach the reference values as resolution increases. At the finest tested grid, with an overall resolution of 200, relative errors in the reported energies and local quantities are generally below 1%: magnetic energy error is 0.46%, kinetic energy error is 0.19%, and drift-frequency error is 0.70%. For case P8 at the same resolution, errors for the listed quantities are no greater than 0.85%; magnetic and kinetic energy errors are 0.05% and 0.23%, respectively. These results are comparable to published local-method results and benchmark spectral solutions. The P8 simulations also reproduce the quasi-steady and oscillatory behaviours associated with different initial magnetic-field strengths. Parallel tests on the Sunway TaihuLight supercomputer scale to 10,368 processor cores. In the strong-scaling tests, the fixed mesh contains about 20.9 million cells. The velocity–temperature–magnetic-field system requires three GMRES iterations across the tested core counts, while its solve time decreases from 6.65 seconds at 1,296 cores to 0.92 seconds at 10,368 cores. For the pressure systems, the two-level preconditioner reduces the iteration count from roughly 145–158 with the one-level method to about 22, cutting solve time by approximately 20–47%. In the weak-scaling tests, each core handles 20 × 20 × 20 cells, and the mesh grows to about 82.9 million cells at 10,368 cores. The two-level method’s pressure-solve time increases by 53% from 648 to 10,368 cores, compared with 139% for the one-level method. However, coarse-grid solves remain the main bottleneck and limit scalability at high core counts. The study demonstrates an accurate and scalable approach to dynamo simulations with pseudo-vacuum boundaries, but it does not provide a complete treatment of planetary-core boundary physics. The authors identify implementing insulating magnetic boundaries and addressing the inner-core singularity as future work. They also note that finer resolutions may require additional multigrid levels, with their effectiveness depending on problem size and communication costs.

Was this synthesis useful?

Score: 0

Why do we need votes?

Votes are used to determine whether we need to re-run our summarizing tools. If the count reaches -10, our tools can be restarted.

Figures and plots

Ask about a figure by number or use its question button to request visual analysis. Up to two figures and four image panels per question. Values read from plots are approximate; check the original data.

AI assistant instructions?

Tools for this paper

Explain it differently 1 free rewrite left today
Save to a collection Group papers in collections to compare them, chat with them and export them. Continue with Google Create a free account Log in

Advanced tools Premium

For researchers who follow a field closely: citation alerts, journal-club slides, version comparison and Zotero export. Discover Premium

Zotero Zotero guide
Citation alerts 3 papers already cite it: be emailed each new one. Citation alerts guide
Who works on it?

Among the 25 papers around this one, the name that comes back most: Andrey Sheyko (3 papers)

9 more people sign several of them. The full map adds their institutions and approaches, from the papers of the last years.

Continue with Google Create a free account: your first map is free
Weekly letter The latest letter of its field, "Physics of matter and light this week" of 16 November 2026: Physics of matter and light this week: Building stability, finding hidden order · Receive one by email for free
Versions

From v1 to v2: Major Version 2 adds a convergence study reporting second-order accuracy in both space and time for velocity, temperature and magnetic field. Spatial errors use a 128-by-128-by-128-by-6 reference mesh at time 0.001, while temporal errors use a fixed 64-by-64-by-64-by-6 mesh and a reference time step of 0.003125 at time 1.

7 more changes, and whether the main results changed, with Premium.

What changed between versions Versions guide

Papers around this one

See them on a map

Ranked by how close their title and abstract are to this paper’s. Select some to read them together.

Closest papers

  1. Very close* Same question Cited by this paper Used in depth by this paper A method it uses
    A spherical shell numerical dynamo benchmark with pseudo-vacuum magnetic boundary conditions
    A. Jackson, Andrey Sheyko, P. Marti et al. · 2014 · Geophysical Journal International · cited 39 times · Not summarized yet

    A spherical-shell dynamo benchmark specifies the pseudo-vacuum boundary conditions used to validate the paper’s dynamo simulations.

  2. Very close* Same question Cited by this paper Used in depth by this paper A method it uses
    A finite-volume solution method for thermal convection and dynamo problems in spherical shells
    H. Harder, U. Hansen · 2005 · Geophysical Journal International · cited 76 times · Not summarized yet

    A finite-volume method for spherical-shell convection and dynamos addresses the same geometry and simulation problem.

  3. Very close Another approach to the problem Cited by this paper
    Parallel and fully implicit simulations of the thermal convection in the Earth’s outer core
    L. Yin, Chao Yang, Shi-Zhuang Ma et al. · 2019 · Computers & Fluids · cited 3 times · Not summarized yet

    This work simulates rotating-shell thermal convection with a fully implicit finite-volume method, rather than the paper’s dynamo-focused approximate-factorization scheme.

    Abstract

    Abstract A parallel and fully implicit method is developed for numerical simulations of the thermal convection of an incompressible fluid in a rotating spherical shell. The method is based on a pseudo-compressibility approach with dual time stepping in order to tackle with the numerical difficulty of the solenoidal condition in the incompressible Navier-Stokes equations. The numerical solution of the nonlinear governing equations is based on the method of lines, whereby the partial differential equations are transformed into ordinary differential equations by a suitable spatial discretization. A second-order cell-centered finite volume discretization based on a collocated cubed-sphere grid i…

  4. Close Same method, another problem Cited by this paper Used in depth by this paper A method it uses
    Applications of a finite-volume algorithm for incompressible MHD problems
    S. Vantieghem, Andrey Sheyko, A. Jackson · 2016 · Geophysical Journal International · cited 13 times · Not summarized yet

    This work applies parallel finite volumes to incompressible MHD and introduces pseudo-pressure to enforce magnetic solenoidality.

    Abstract

    We present the theory, algorithms and implementation of a parallel finite-volume algorithm for the solution of the incompressible magnetohydrodynamic (MHD) equations using unstructured grids that are applicable for a wide variety of geometries. Our method implements a mixed Adams-Bashforth/Crank-Nicolson scheme for the nonlinear terms in the MHD equations and we prove that it is stable independent of the time step. To ensure that the solenoidal condition is met for the magnetic field, we use a method whereby a pseudo-pressure is introduced into the induction equation; since we are concerned with incompressible flows, the resulting Poisson equation for the pseudo-pressure is solved alongside …

  5. Close* Same question Cited by this paper
    Numerical investigations of rotating MHD in a spherical shell
    Andrey Sheyko · 2014 · cited 16 times · Not summarized yet

    This study investigates rotating MHD in a spherical shell, the same physical setting as the paper’s dynamo simulations.

  6. Close* Broader context Cited by this paper
    Magnetohydrodynamics in Rapidly Rotating spherical Systems
    K. Zhang, G. Schubert · 2000 · Annual Review of Fluid Mechanics · cited 116 times · Not summarized yet

    This work addresses magnetohydrodynamics in rapidly rotating spherical systems, the broader physical setting of the paper.

  7. Close Another approach to the problem
    Finite Element Approximation of Rotating MHD-Boussinesq Flows with Nonlinear Power-Law Viscosity
    A. Panda, Evana Islam Sarkar · 2026 · Not summarized yet

    This work approximates rotating MHD-Boussinesq flows with conforming finite elements and nonlinear power-law viscosity.

    Abstract

    We propose and analyse a fully discrete finite element approximation of the three-dimensional incompressible rotating MHD-Boussinesq system with nonlinear power-law ($p$-Laplace) viscosity. Conforming elements are used for the velocity and temperature, $\mathbf H(\operatorname{curl})$-conforming N\'ed\'elec elements for the magnetic induction, and a semi-implicit backward Euler scheme in time. The discretisation preserves the principal energy structure: the Lorentz and induction terms cancel exactly, while the Coriolis force remains energy neutral. This yields well-posedness and mesh-independent energy stability. The nonlinear power-law diffusion creates two main analytical difficulties. Sin…

  8. Close Same question Cited by this paper A method it uses
    Systematic parameter study of dynamo bifurcations in geodynamo simulations
    L. Petitdemange · 2018 · Physics of the Earth and Planetary Interiors · cited 19 times · Not summarized yet

    This study examines dynamo bifurcations in thermally driven rotating spherical-shell simulations, rather than numerical scalability.

    Abstract

    Abstract We investigate the nature of the dynamo bifurcation in a configuration applicable to the Earth’s liquid outer core, i.e. in a rotating spherical shell with thermally driven motions with no-slip boundaries. Unlike in previous studies on dynamo bifurcations, the control parameters have been varied significantly in order to deduce general tendencies. Numerical studies on the stability domain of dipolar magnetic fields found a dichotomy between non-reversing dipole-dominated dynamos and the reversing non-dipole-dominated multipolar solutions. We show that, by considering weak initial fields, the above transition disappears and is replaced by a region of bistability for which dipolar and…

Prior work it builds on

  1. Close* Same method, another problem Used in depth by this paper A method it uses
    Parallel numerical simulation of the thermal convection in the Earth’s outer core on the cubed-sphere
    L. Yin, Chao Yang, Shi-Zhuang Ma et al. · 2017 · Geophysical Journal International · cited 3 times · Not summarized yet

    This work parallelizes cubed-sphere simulations of outer-core thermal convection, sharing the paper’s grid geometry and parallel setting.

  2. Related* Same question Used in depth by this paper A method it uses
    A nonlinear vacillating dynamo induced by an electrically heterogeneous mantle
    Kit H. Chan, K. Zhang, J. Zou et al. · 2001 · Geophysical Research Letters · cited 13 times · Not summarized yet

    This study examines a vacillating dynamo in a planetary setting, addressing dynamo behavior rather than solver performance.

  3. Related* Another approach to the problem Used in depth by this paper A method it uses
    An integro-differential formulation for magnetic induction in bounded domains: boundary element-finite volume method
    A. Iskakov, S. Descombes, E. Dormy · 2004 · Journal of Computational Physics · cited 49 times · Not summarized yet

    This work models magnetic induction in bounded domains with a boundary-element and finite-volume formulation.

  4. Close Review
    Theory and Modeling of Planetary Dynamos
    Johannes Wicht, A. Tilgner · 2010 · Space Science Reviews · cited 62 times · Not summarized yet

    This article reviews planetary dynamo equations, numerical methods, model variations, and parameter-dependent regimes.

    Abstract

    Numerical dynamo models are increasingly successful in modeling many features of the geomagnetic field. Moreover, they have proven to be a useful tool for understanding how the observations connect to the dynamo mechanism. More recently, dynamo simulations have also ventured to explain the surprising diversity of planetary fields found in our solar system. Here, we describe the underlying model equations, concentrating on the Boussinesq approximations, briefly discuss the numerical methods, and give an overview of existing model variations. We explain how the solutions depend on the model parameters and introduce the primary dynamo regimes. Of particular interest is the dependence on the Ekm…

  5. Close* Same method, another problem A method it uses
    A non-staggered, conservative, V×B=0' finite-volume scheme for 3D implicit extended magnetohydrodynamics in curvilinear geometries
    L. Chacón · 2004 · Computer Physics Communications · cited 85 times · Not summarized yet

    This work develops a conservative finite-volume MHD scheme for curvilinear geometries, sharing the paper’s discretization family.

Papers that cite it

  1. Close* Builds on this paper
    Effect of an external axial magnetic field on thermal convection of an electric conducting fluid in a rotating spherical shell
    Hidemoto Satake, T. Tagawa · 2024 · International Communications in Heat and Mass Transfer · cited 1 time · Not summarized yet

    This study cites the paper while examining how an external axial magnetic field affects convection in a rotating spherical shell.

  2. Related* Broader context
    Optimization design and heat performance analysis of a cylindrical wind-magnetic eddy current heater
    Hong-Juan Cheng, Qing Yang, Yun Ding et al. · 2025 · International journal of thermal sciences · cited 3 times · Not summarized yet

    This study cites the paper in an analysis of a cylindrical wind-magnetic heater, a neighboring application of magnetic-fluid modeling.

How are these papers found?

We compare the title and abstract of this paper with those of every paper summarized on this site, and with the papers Semantic Scholar recommends, those it cites and those citing it (for a much-cited paper, its most recent citations and, from OpenAlex, the most cited ones). Among the prior work and the follow-ups, the papers that use each other in depth, a method it uses and the much-cited follow-ups come a little higher. The levels (very close, close, related) come from the similarity of the texts, not from a judgment of quality; * marks a paper without an abstract, compared by its title. The sentence under a paper is written by an AI from the two abstracts: check it in the papers. A paper marked not relevant by three readers leaves the list.

Updated on 19 November 2026.

The illustrated guide to the papers around a paper

Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.