A5073696890- Parallel Computing and Optimization Techniques
- Distributed and Parallel Computing Systems
- Advanced Numerical Methods in Computational Mathematics
- Advanced Data Storage Technologies
- Scientific Computing and Data Management
- Matrix Theory and Algorithms
- Elasticity and Material Modeling
- Numerical methods for differential equations
- Computational Physics and Python Applications
- Cryospheric studies and observations
- Superconducting Materials and Applications
- Computational Fluid Dynamics and Aerodynamics
- Geological Modeling and Analysis
- Meteorological Phenomena and Simulations
- Arctic and Antarctic ice dynamics
- Model Reduction and Neural Networks
- Landslides and related hazards
- Radiomics and Machine Learning in Medical Imaging
- Electromagnetic Simulation and Numerical Methods
- Medical Imaging Techniques and Applications
- Plasma and Flow Control in Aerodynamics
- Fluid Dynamics and Turbulent Flows
- Particle accelerators and beam dynamics
- Numerical Methods and Algorithms
- Magnetic confinement fusion research
University of Colorado Boulder
2016-2025
Argonne National Laboratory
2012-2024
University of Colorado System
2015-2023
Lawrence Livermore National Laboratory
2017-2021
Memorial University of Newfoundland
2019
Rice University
2017
Intel (United States)
2017
Applied Mathematics (United States)
2017
University of Chicago
2016
Lawrence Berkeley National Laboratory
2016
The shallow shelf approximation is a better ``sliding law'' for ice sheet modeling than those sliding laws in which basal velocity function of driving stress. as formulated by \emph{Schoof} [2006a] well-suited to this use. Our new thermomechanically coupled scheme based on plasticity assumption about the strength saturated till underlying yield stress given Mohr-Coulomb formula using modeled pore water pressure. Using scheme, our prognostic whole model has convincing streams. Driving...
We used time‐lapse imagery, seismic and audio recordings, iceberg glacier velocities, ocean wave measurements, simple theoretical considerations to investigate the interactions between Jakobshavn Isbræ its proglacial ice mélange. The mélange behaves as a weak, granular shelf whose rheology varies seasonally. Sea growth in winter stiffens matrix by binding clasts together, ultimately preventing calving of full‐glacier‐thickness icebergs (the dominant style calving) enabling several kilometer...
We consider multiphysics applications from algorithmic and architectural perspectives, where “algorithmic” includes both mathematical analysis computational complexity, “architectural” software hardware environments. Many diverse can be reduced, en route to their simulation, a common algebraic coupling paradigm. Mathematical of in this form is not always practical for realistic applications, but model problems representative discussed herein provide insight. A variety frameworks have been...
High-quality ordinary differential equation (ODE) solver libraries have a long history, going back to the 1970s. Over past several years we implemented, on top of PETSc linear and nonlinear package, new general-purpose, extensive, extensible library for solving ODEs algebraic equations (DAEs). Package includes support both forward adjoint sensitivities that can be easily utilized by TAO optimization which is also part PETSc. The ODE/DAE integrator strives highly scalable but deliver high...
Simulations of long-term lithospheric deformation involve post-failure analysis high-contrast brittle materials driven by buoyancy and processes at the free surface. Geodynamic phenomena such as subduction continental rifting take place over millions year time scales, thus require efficient solution methods. We present pTatin3D, a geodynamics modeling package utilising material-point-method for tracking material composition, combined with multigrid finite-element method to solve...
Abstract The model used by Lingle and Clark (1985) to approximate the deformation of Earth under a single ice stream is adapted purposes continent-scale ice-sheet simulation. combines layered elastic spherical (Farrell, 1972) with viscous half-space overlain an plate lithosphere (Cathles, 1975). For we identify new mathematical formulation, essentially time-dependent partial differential equation, which generalizes improves upon standard relaxing asthenosphere widely in formulation allows...
Performance tests and analyses are critical to effective high-performance computing software development central components in the design implementation of computational algorithms for achieving faster simulations on existing future architectures large-scale application problems. In this article, we explore performance space-time trade-offs important compute-intensive kernels numerical solvers partial differential equations (PDEs) that govern a wide range physical applications. We consider...
Abstract Comparison of numerically computed solutions to exact (analytical) time-dependent solutions, when possible, is superior intercomparison as a technique for verification numerical models. At least two sources such exist the isothermal shallow ice-sheet equation: similarity and with ‘compensatory accumulation’. In this paper, we derive new non-zero accumulation. We also (i) sinusoidal-in-time accumulation (ii) basal sliding. A specific test suite based on these proposed used verify...
This three section report can be regarded as an extended appendix to (Bueler, Brown, and Lingle 2006). First we give the detailed construction of exact solution a standard continuum model cold, shallow, thermocoupled ice sheet. The is by calculation compensatory accumulation heat source functions which make chosen pair for thickness temperature into solutions coupled system. construct here ``TestG'' in (Bueler others, 2006) steady state ``Test F'' special case. In second reference C...
The Portable, Extensible Toolkit for Scientific computing (PETSc), which focuses on the scalable solution of problems based partial differential equations, now incorporates new components that allow full compos ability solvers multiphysics and multilevel methods. Through strong encapsulation, we achieve arbitrary, dynamic composition hierarchical methods coupled customization all in composite solvers. For example, support block decompositions with nested multigrid as well fully system...
Efficient exploitation of exascale architectures requires rethinking the numerical algorithms used in many large-scale applications. These favor that expose ultra fine-grain parallelism and maximize ratio floating point operations to energy intensive data movement. One few viable approaches achieve high efficiency area PDE discretizations on unstructured grids is use matrix-free/partially-assembled high-order finite element methods, since these methods can increase accuracy and/or lower...
The Portable Extensible Toolkit for Scientific Computation (PETSc) library provides scalable solvers nonlinear time-dependent differential and algebraic equations numerical optimization via the Advanced Optimization (TAO). PETSc is used in dozens of scientific fields an important building block many simulation codes. During U.S. Department Energy’s Exascale Computing Project, team has made substantial efforts to enable efficient utilization massive fine-grain parallelism present within...
This presentation focuses on numerical solutions for Initial Value Problems (IVPs) involving linear PDEs dominating the time step size, as is case dynamical cores. We investigate using Rational Approximation of Exponential Integration (REXI). REXI replaces sequential time-stepping with a sum rational terms, enabling parallelization and exploiting additional scalability supercomputers spatially limited problems.We introduce "unified REXI" method, showing its algebraic equivalence to methods...
PetscSF, the communication component of Portable, Extensible Toolkit for Scientific Computation (PETSc), is designed to provide PETSc's infrastructure suitable exascale computers that utilize GPUs and other accelerators. PetscSF provides a simple application programming interface (API) managing common patterns in scientific computations by using star-forest graph representation. supports several implementations based on MPI NVSHMEM, whose selection characteristics or target architecture. An...