Thornado-transport: Anderson- And GPU-accelerated nonlinear solvers for neutrino-matter coupling

M. Paul Laiu; J. Austin Harris; Ran Chu; Eirik Endeve

doi:10.1088/1742-6596/1623/1/012013

Thornado-transport: Anderson- And GPU-accelerated nonlinear solvers for neutrino-matter coupling

M. Paul Laiu, J. Austin Harris, Ran Chu, Eirik Endeve

Research output: Contribution to journal › Conference article › peer-review

4 Scopus citations

Abstract

Algorithms for neutrino-matter coupling in core-collapse supernovae (CCSNe) are investigated in the context of a spectral two-moment model, which is discretized in space with the discontinuous Galerkin method, integrated in time with implicit-explicit (IMEX) methods, and implemented in the toolkit for high-order neutrino-radiation hydrodynamics (thornado). The model considers electron neutrinos and antineutrinos and tabulated opacities from Bruenn (1985), which includes neutrino-electron scattering and pair processes. The nonlinear system arising from implicit time discretization of the equations governing neutrino-matter coupling is iterated to convergence using Anderson-accelerated fixed-point methods, which avoid formation of Jacobians and inversion of dense linear systems. Numerical experiments show that, for a given tolerance, a nested iteration scheme which aims to reduce opacity evaluations can lower the computational cost. Our initial port to GPUs, using both OpenMP and OpenACC, shows an overall speedup of up to ∼ 100 when compared to results using a single CPU core. These results indicate that the algorithms implemented in thornado are well-suited to GPU acceleration.

Original language	English
Article number	012013
Journal	Journal of Physics: Conference Series
Volume	1623
Issue number	1
DOIs	https://doi.org/10.1088/1742-6596/1623/1/012013
State	Published - Sep 24 2020
Event	14th International Conference on Numerical Modeling of Space Plasma Flows, ASTRONUM 2019 - Paris, France Duration: Jul 1 2019 → Jul 5 2019

Funding

This research used resources of the Oak Ridge Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC05-00OR22725. This research was sponsored, in part, by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory (ORNL), managed by UT-Battelle, LLC. This research was also supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. Department of Energy Office of Science and the National Nuclear Security Administration. thornado uses equation of state and opacity tables and subroutines developed under the WeakLib project. We acknowledge the contributions to WeakLib of Ryan Landfield and Eric J. Lentz. 1 This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

Funders	Funder number
DOE Office of Science	DE-AC05-00OR22725
U.S. Department of Energy Office of Science
UT-Battelle	17-SC-20-SC
U.S. Department of Energy
National Nuclear Security Administration
Oak Ridge National Laboratory

Access to Document

10.1088/1742-6596/1623/1/012013

Cite this

@article{49fe387b585f433c9b95f76c9c8ab1ad,

title = "Thornado-transport: Anderson- And GPU-accelerated nonlinear solvers for neutrino-matter coupling",

abstract = "Algorithms for neutrino-matter coupling in core-collapse supernovae (CCSNe) are investigated in the context of a spectral two-moment model, which is discretized in space with the discontinuous Galerkin method, integrated in time with implicit-explicit (IMEX) methods, and implemented in the toolkit for high-order neutrino-radiation hydrodynamics (thornado). The model considers electron neutrinos and antineutrinos and tabulated opacities from Bruenn (1985), which includes neutrino-electron scattering and pair processes. The nonlinear system arising from implicit time discretization of the equations governing neutrino-matter coupling is iterated to convergence using Anderson-accelerated fixed-point methods, which avoid formation of Jacobians and inversion of dense linear systems. Numerical experiments show that, for a given tolerance, a nested iteration scheme which aims to reduce opacity evaluations can lower the computational cost. Our initial port to GPUs, using both OpenMP and OpenACC, shows an overall speedup of up to ∼ 100 when compared to results using a single CPU core. These results indicate that the algorithms implemented in thornado are well-suited to GPU acceleration.",

author = "{Paul Laiu}, M. and Harris, {J. Austin} and Ran Chu and Eirik Endeve",

note = "Publisher Copyright: {\textcopyright} Published under licence by IOP Publishing Ltd.; 14th International Conference on Numerical Modeling of Space Plasma Flows, ASTRONUM 2019 ; Conference date: 01-07-2019 Through 05-07-2019",

year = "2020",

month = sep,

day = "24",

doi = "10.1088/1742-6596/1623/1/012013",

language = "English",

volume = "1623",

journal = "Journal of Physics: Conference Series",

issn = "1742-6588",

number = "1",

}

TY - JOUR

T1 - Thornado-transport

T2 - 14th International Conference on Numerical Modeling of Space Plasma Flows, ASTRONUM 2019

AU - Paul Laiu, M.

AU - Harris, J. Austin

AU - Chu, Ran

AU - Endeve, Eirik

PY - 2020/9/24

Y1 - 2020/9/24

N2 - Algorithms for neutrino-matter coupling in core-collapse supernovae (CCSNe) are investigated in the context of a spectral two-moment model, which is discretized in space with the discontinuous Galerkin method, integrated in time with implicit-explicit (IMEX) methods, and implemented in the toolkit for high-order neutrino-radiation hydrodynamics (thornado). The model considers electron neutrinos and antineutrinos and tabulated opacities from Bruenn (1985), which includes neutrino-electron scattering and pair processes. The nonlinear system arising from implicit time discretization of the equations governing neutrino-matter coupling is iterated to convergence using Anderson-accelerated fixed-point methods, which avoid formation of Jacobians and inversion of dense linear systems. Numerical experiments show that, for a given tolerance, a nested iteration scheme which aims to reduce opacity evaluations can lower the computational cost. Our initial port to GPUs, using both OpenMP and OpenACC, shows an overall speedup of up to ∼ 100 when compared to results using a single CPU core. These results indicate that the algorithms implemented in thornado are well-suited to GPU acceleration.

AB - Algorithms for neutrino-matter coupling in core-collapse supernovae (CCSNe) are investigated in the context of a spectral two-moment model, which is discretized in space with the discontinuous Galerkin method, integrated in time with implicit-explicit (IMEX) methods, and implemented in the toolkit for high-order neutrino-radiation hydrodynamics (thornado). The model considers electron neutrinos and antineutrinos and tabulated opacities from Bruenn (1985), which includes neutrino-electron scattering and pair processes. The nonlinear system arising from implicit time discretization of the equations governing neutrino-matter coupling is iterated to convergence using Anderson-accelerated fixed-point methods, which avoid formation of Jacobians and inversion of dense linear systems. Numerical experiments show that, for a given tolerance, a nested iteration scheme which aims to reduce opacity evaluations can lower the computational cost. Our initial port to GPUs, using both OpenMP and OpenACC, shows an overall speedup of up to ∼ 100 when compared to results using a single CPU core. These results indicate that the algorithms implemented in thornado are well-suited to GPU acceleration.

UR - http://www.scopus.com/inward/record.url?scp=85092789781&partnerID=8YFLogxK

U2 - 10.1088/1742-6596/1623/1/012013

DO - 10.1088/1742-6596/1623/1/012013

M3 - Conference article

AN - SCOPUS:85092789781

SN - 1742-6588

VL - 1623

JO - Journal of Physics: Conference Series

JF - Journal of Physics: Conference Series

IS - 1

M1 - 012013

Y2 - 1 July 2019 through 5 July 2019

ER -

Thornado-transport: Anderson- And GPU-accelerated nonlinear solvers for neutrino-matter coupling

Abstract

Funding

Access to Document

Other files and links

Fingerprint

Cite this