Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures
Abstract
The Landau collision integral is an accurate model for the smallangle dominated Coulomb collisions in fusion plasmas. Here, we investigate a high order accurate, fully conservative, finite element discretization of the nonlinear multispecies Landau integral with adaptive mesh refinement using the PETSc library (ulwww.mcs.anl.gov/petsc). We develop algorithms and techniques to efficiently utilize emerging architectures with an approach that minimizes memory usage and movement and is suitable for vector processing. The Landau collision integral is vectorized with Intel AVX512 intrinsics and the solver sustains as much as 22% of the theoretical peak flop rate of the Second Generation Intel Xeon Phi (``Knights Landing'') processor.
 Authors:

 Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States). Scalable Solvers Group
 Princeton Plasma Physics Lab. (PPPL), Princeton, NJ (United States)
 Rice Univ., Houston, TX (United States). Computational and Applied Mathematics
 Univ. of Colorado, Boulder, CO (United States). Dept. of Computer Science
 Intel Corp., Santa Clara, CA (United States)
 Publication Date:
 Research Org.:
 Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Argonne National Lab. (ANL), Argonne, IL (United States)
 Sponsoring Org.:
 USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC21)
 OSTI Identifier:
 1465410
 Alternate Identifier(s):
 OSTI ID: 1473684
 Grant/Contract Number:
 AC0205CH11231; AC0209CH11466; AC0206CH11357
 Resource Type:
 Accepted Manuscript
 Journal Name:
 SIAM Journal on Scientific Computing
 Additional Journal Information:
 Journal Volume: 39; Journal Issue: 6; Journal ID: ISSN 10648275
 Publisher:
 SIAM
 Country of Publication:
 United States
 Language:
 English
 Subject:
 97 MATHEMATICS AND COMPUTING; Landau collision integral; fusion plasma physics
Citation Formats
Adams, Mark F., Hirvijoki, Eero, Knepley, Matthew G., Brown, Jed, Isaac, Tobin, and Mills, Richard. Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures. United States: N. p., 2017.
Web. doi:10.1137/17M1118828.
Adams, Mark F., Hirvijoki, Eero, Knepley, Matthew G., Brown, Jed, Isaac, Tobin, & Mills, Richard. Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures. United States. doi:10.1137/17M1118828.
Adams, Mark F., Hirvijoki, Eero, Knepley, Matthew G., Brown, Jed, Isaac, Tobin, and Mills, Richard. Wed .
"Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures". United States. doi:10.1137/17M1118828. https://www.osti.gov/servlets/purl/1465410.
@article{osti_1465410,
title = {Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures},
author = {Adams, Mark F. and Hirvijoki, Eero and Knepley, Matthew G. and Brown, Jed and Isaac, Tobin and Mills, Richard},
abstractNote = {The Landau collision integral is an accurate model for the smallangle dominated Coulomb collisions in fusion plasmas. Here, we investigate a high order accurate, fully conservative, finite element discretization of the nonlinear multispecies Landau integral with adaptive mesh refinement using the PETSc library (ulwww.mcs.anl.gov/petsc). We develop algorithms and techniques to efficiently utilize emerging architectures with an approach that minimizes memory usage and movement and is suitable for vector processing. The Landau collision integral is vectorized with Intel AVX512 intrinsics and the solver sustains as much as 22% of the theoretical peak flop rate of the Second Generation Intel Xeon Phi (``Knights Landing'') processor.},
doi = {10.1137/17M1118828},
journal = {SIAM Journal on Scientific Computing},
number = 6,
volume = 39,
place = {United States},
year = {2017},
month = {12}
}