Bulletin of the American Physical Society
APS April Meeting 2019
Volume 64, Number 3
Saturday–Tuesday, April 13–16, 2019; Denver, Colorado
Session G11: Numerical Relativity: Algorithms and Code Development |
Hide Abstracts |
Sponsoring Units: DGRAV Chair: Thomas Baumgarte, Bowdoin College Room: Sheraton Governor's Square 17 |
Sunday, April 14, 2019 8:30AM - 8:42AM |
G11.00001: SpECTRE: A task-based discontinuous Galerkin code for relativistic astrophysics Lawrence E Kidder We provide an update on the development of SpECTRE (https://github.com/sxs-collaboration/spectre), a new open-source relativistic astrophysics code that combines a discontinuous Galerkin method with a task-based parallelism model. SpECTRE's goal is to achieve more accurate solutions for challenging relativistic astrophysics problems such as core-collapse supernovae and binary neutron star mergers, while making efficient use of the largest supercomputers. |
Sunday, April 14, 2019 8:42AM - 8:54AM |
G11.00002: Category Theory for HPC Software Engineering Erik Schnetter Most, if not all, current numerical relativity codes model general relativistic concepts at a very low level: Concepts such as "manifold", "vector space", or "basis" are not rigorously defined in the code and are only expressed via low-level mechanisms such as "array", "loop", or "index". This makes it tedious to express physics concepts (such as the Einstein equations) in a code, and implementations then fail to be sufficiently generic. At the same time, modern computing hardware is becoming more complex, and respective necessary code optimizations are difficult to express, further complicating the code and obscuring the physics. I will give examples how certain concepts "stolen" from category theory can improve both code clarity and code performance, and will argue for a higher mathematical abstraction level in numerical software. |
Sunday, April 14, 2019 8:54AM - 9:06AM |
G11.00003: Dendro-GR: A scalable framework for Adaptive Computational General Relativity on Heterogeneous Clusters David W Neilsen, Milinda Fernando, Hari Sundar, Eric Winston Hirschmann We present a portable and highly-scalable frame-work that targets problems in the astrophysics and numerical relativity communities. This framework combines together the parallel DENDRO octree with wavelet adaptive multiresolution and a automatic code-generation physics module to solve the Ein- stein equations of general relativity in the BSSNOK formulation. The goal of this work is to perform advanced, massively parallel numerical simulations of binary black hole and neutron star mergers, including Intermediate Mass Ratio Inspirals (IMRIs) of binary black holes with mass ratios on the order of 100:1. We will discuss the development of automatic code generators for computational relativity supporting SIMD vectorization, OpenMP, and CUDA combined with efficient distributed memory adaptive data-structures. Preliminary results of binary black hole mergers will be presented. |
Sunday, April 14, 2019 9:06AM - 9:18AM |
G11.00004: Efficient extraction of Weyl scalars from numerical simulations using a non-orthonormal tetrad Dante A B Iozzo, Michael Boyle, Mark A Scheel, Lawrence E Kidder, Saul A Teukolsky We present a procedure that numerically constructs the Weyl scalars from the characteristic fields of the Weyl tensor evolution system. These characteristic fields are spatial, real, and computed directly from the extrinsic curvature and spatial Ricci tensor, which leads to efficient computation. Further computational efficiency improvements can be made by using a non-orthonormal tetrad defined as in Minkowski space and using an extrapolation procedure. An analysis of the extraction procedure is presented and the possible issue of junk radiation backscattering is addressed. |
Sunday, April 14, 2019 9:18AM - 9:30AM |
G11.00005: Ameliorating the Courant Limitations of Spherical Coordinates in Numerical Simulations of the Einstein Equations Yosef Zlochower, Vassilios Mewes, Manuela Campanelli Explicit schemes for the evolutions of hyperbolic systems in spherical coordinates have the well known Courant (CFL) limitation that the maximum stable timestep scales as dt ~ dr dtheta dphi. In this talk, we compare a variety of methods we used to significantly increase the size of the timestep BSSN simulations using spherical coordinates while maintaining high accuracy. For our methods, the timestep scales as dt ~ dr. |
Sunday, April 14, 2019 9:30AM - 9:42AM |
G11.00006: Hyperbolic Initial Data for Boosted and Spinning Black Holes Maria C Babiuc, Jeffrey H Winicour The two-body problem in general relativity -- such as black hole mergers -- plays an important role in decoding the gravitational wave signals. This problem is unsolved analytically and is numerically challenging, because although when given correct initial data, the Einstein equations will yield the expected solution, they are very sensitive to numerical errors and could lead to wrong spacetimes. Standard techniques for constructing initial data use elliptic equations that require inner boundary conditions and are prone to `junk' radiation. We present work in progress to numerically implement and test a hyperbolic-algebraic method to solve the initial data constraints. This method, proposed by I. Racz, does not rely on conformal flatness and is mathematically well posed. We develop HyperSolID, a code that calculates Hyperbolic Solutions to Initial Data for black hole spacetimes. The numerical algorithm implements the Hamiltonian and momentum constraints as a system of well-posed, hyperbolic-algebraic equations in stereographic coordinates on a logarithmic grid. We test the code with two trials: a near-light-speed boosted Schwarzchild and high-spin Kerr black hole, in order to asses its performance with nontrivial spacetimes. |
Sunday, April 14, 2019 9:42AM - 9:54AM |
G11.00007: Neutron star evolutions with WENO discontinuous Galerkin methods Francois Hebert Discontinuous Galerkin methods combine high accuracy with shock-handling properties. This combination makes them well-suited to solve the PDEs describing relativistic astrophysical scenarios such as neutron star mergers and accretion discs, where fluid shocks and surfaces naturally arise. However, the development of schemes for handling shocks in a discontinuous Galerkin code is ongoing, with recent efforts working toward increasing the robustness and accuracy of these schemes. Using the SpECTRE (https://github.com/sxs-collaboration/spectre) code, we compare different shock-handling schemes: simple slope limiters and more recent, higher-order WENO-based limiters. We evaluate the performance of these different limiters for standard shock problems and neutron star evolutions. |
Sunday, April 14, 2019 9:54AM - 10:06AM |
G11.00008: A no boundary method for numerical relativity David Garfinkle, Lydia Bieri Numerical relativity methods either have an outer boundary at infinity or use an unsuitable outer boundary condition at a finite outer boundary. We present a method that has a finite outer boundary, but has no need for an outer boundary condition. This is done by having a part of the initial data on a spacelike but outwardly expanding surface. |
Sunday, April 14, 2019 10:06AM - 10:18AM |
G11.00009: Local time-stepping in discontinuous Galerkin GRHMD evolution William T Throwe Simulations involving a large range of length or time scales are inefficient due to the number of evaluations of evolution equations being set by the most stringent scales. Using a local time stepping scheme, where different degrees of freedom are evaluated at different times, can resolve this issue, but such schemes are generally low-order or fail to preserve conservation laws. We present a generic conservative, high-order local time stepping scheme and demonstrate applications to evolution of the GRMHD equations. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2020 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700