Search NASA⌕ Search

SEARCH · Search NASA

Results for “direct solver”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5

Parallelized Characteristic Basis Finite Element Method (CBFEM-MPI)-A non-iterative domain decomposition algorithm for electromagnetic scattering problems

In this paper, we introduce a parallelized version of a novel, non-iterative domain decomposition algorithm, called Characteristic Basis Finite Element Method (CBFEM-MPI), for efficient solution of large-scale electromagnetic scattering problems, by utilizing a set of specially defined characteristic basis functions (CBFs). This approach is based on the decomposition of the computational domain into a number of non-overlapping subdomains wherein the CBFs are generated by employing a novel procedure, which differs from all those that have been used in the past. Clearly, the CBFs are obtained by calculating the fields radiated by a finite number of dipole-type sources, which are placed hypothetically along the boundary of the conducting object. The major advantages of the proposed technique are twofold: (i) it provides a substantial reduction in the matrix size, and thus, makes use of direct solvers efficiently and (ii) it enables the utilization of parallel processing techniques that considerably decrease the overall computation time. We illustrate the application of the proposed approach via several 3D electromagnetic scattering problems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Application of the inverse fast multipole method as a preconditioner in a 3D Helmholtz boundary element method

Here we investigate an efficient preconditioning of iterative methods (such as GMRES) for solving dense linear systems Ax = b that follow from a boundary element method (BEM) for the 3D Helmholtz equation, focusing on the low-frequency regime. While matrix–vector products in GMRES can be accelerated through the low-frequency fast multipole method (LFFMM), the BEM often remains computationally expensive due to the large number of GMRES iterations. We propose the application of the inverse fast multipole method (IFMM) as a preconditioner to accelerate the convergence of GMRES. The IFMM is in essence an approximate direct solver that uses a multilevel hierarchical decomposition and low-rank approximations. The proposed IFMM-based preconditioning has a tunable parameter ε that balances the cost to construct a preconditioner M, which is an approximation of A -1 , and the cost to perform the iterative process by means of M. Namely, using a small (respectively, large) value of ε takes a long (respectively, short) time to construct M, while the number of iterations can be small (respectively, large). A comprehensive set of numerical examples involving various boundary value problems with complicated geometries and mixed boundary conditions is presented to validate the efficiency of the proposed method. We show that the IFMM preconditioner (with a nearly optimal ε of 10 -2 ) clearly outperforms some common preconditioners for the BEM, achieving 1.2–10.8 times speed-up of the computations, in particular when the scale of the underlying scatterer is about five wavelengths or more. In addition, the IFMM preconditioner is capable of solving complicated problems (in a reasonable amount of time) that BD preconditioner can not.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Application of linear multifrequency-grey acceleration to preconditioned Krylov iterations for thermal radiation transport

The thermal radiative transfer (TRT) equations comprise a radiation equation coupled to the material internal energy equation. Linearization of these equations produces effective, thermally-redistributed scattering through absorption-reemission. In this paper, we investigate the effectiveness and efficiency of Linear-Multi-Frequency-Grey (LMFG) acceleration that has been reformulated for use as a preconditioner to Krylov iterative solution methods. We introduce two general frameworks, the scalar flux formulation (SFF) and the absorption rate formulation (ARF), and investigate their iterative properties in the absence and presence of true scattering. SFF has a group-dependent state size but may be formulated without inner iterations in the presence of scattering, while ARF has a group-independent state size but requires inner iterations when scattering is present. We compare and evaluate the computational cost and efficiency of LMFG applied to these two formulations using a direct solver for the preconditioners. Finally, this work is novel because the use of LMFG for the radiation transport equation, in conjunction with Krylov methods, involves special considerations not required for radiation diffusion.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Predicting Flow in Fracture Networks With Quantum Algorithms

Uncertainty quantification plays a crucial role in the modeling of subsurface flow. For instance, uncertainties in the properties of geologic fracture networks significantly impact flow, requiring numerous simulations to accurately estimate quantities of interest. However, each simulation is computationally expensive because it requires solving a large linear system to capture features that involve both small and large fractures. An example is in percolation, where the interaction of many small fractures (which cumulatively can have a large surface area) with the rock matrix must be modeled precisely. Quantum computing is an emerging tool with the potential to address this issue. Quantum algorithms offer a significant speedup in solving linear systems, achieving efficiencies that are challenging to match with classical approaches. These classical approaches include direct solvers, such as LU decomposition, and iterative methods, notably preconditioned conjugate gradient, commonly used in subsurface modeling to solve large sparse systems. However, applying quantum algorithms to geologic fracture flow requires careful attention to algorithmic and problem-specific constraints to fully realize this quantum advantage. In this work we describe a quantum algorithm for generalized Monte Carlo applications with a quadratic speedup over the classical approaches which can be combined with the quantum speedup, currently under investigation, for solving quantum linear systems for subsurface flow. We show that for quantum algorithms the computational cost of estimating a quantity of interest for a statistical ensemble of networks is roughly the same as that of a single realization, essentially implying that one can get uncertainty quantification for free.

58 GEOSCIENCES↗

Hybrid iterative approach for simulation of radio-frequency fields in plasma

In this work, a novel iterative approach for solving discretized linear wave equations in a frequency domain, which combines time evolution with iterative relaxation schemes, is presented. In this hybrid approach, each iteration cycle consists of evolution of electromagnetic (EM) fields in time over a specified number of field periods followed by several iterative relaxations. Provided that there is sufficient dissipation, both the time evolution and the iterative relaxations contribute to the convergence of the EM fields to the solution of the formulated full wave boundary value problem. Time evolution rapidly distributes EM fields, propagating with group velocity, over the simulation domain, while the iterative relaxations smooth the fields, reducing the numerical errors such that iteration cycles converge to a steady state solution, approximating the solution of the formulated problem. This approach is intended for large scale simulations which are beyond the capabilities of direct solvers presently used for solving wave equations in the frequency domain. Lastly, the technique is demonstrated for solving wave equations on a regular grid using a cold plasma dielectric model with collisions for 2D modeling of EM fields in tokamak in an electron cyclotron frequency range.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A WKB based preconditioner for the 1D Helmholtz wave equation

Frequency-domain full-wave solutions to the cold-plasma problem have become ubiquitous in the study of radio frequency power in fusion plasmas. However, recent efforts at extreme levels of geometric fidelity have revealed fundamental limits in the problem size that can be solved by typical sparse direct solver based methods. These limits are of particular importance in the 3D study of RF launchers, where the number of degrees of freedom required can exceed 100 million. In such cases, it would be advantageous to solve the system via iterative means, but due to the large null space of the curl-curl operator, the convergence properties of algorithms like GMRES are poor. Here we present a physics-based preconditioner in the form of a WKB solution and demonstrate the iterative solution to the frequency-domain Helmholtz problem in 1D for several cases ranging from satisfying the WKB approximation to strongly violating it.

Green, David↗

Linear complexity

We present factorization and solution phases for a new linear complexity direct solver designed for concurrent batch operations on fine-grained parallel architectures, for matrices amenable to hierarchical representation. We focus on the strong-admissibility-based $\mathscr{H}^{2}$ format, where strong recursive skeletonization factorization compresses remote interactions. We build upon previous implementations of $\mathscr{H}^{2}$ matrix construction for efficient factorization and solution algorithm design, which are illustrated graphically in stepwise detail. The algorithms are ‘blackbox’ in the sense that the only inputs are the matrix and right-hand side, without analytical or geometrical information about the origin of the system. We demonstrate linear complexity scaling in both time and memory on four representative families of dense matrices up to one million in size. Parallel scaling up to 16 threads is enabled by a multi-level matrix graph coloring and avoidance of dynamic memory allocations thanks to prefix-sum memory management. An experimental backward error analysis is included. We break down the timings of different phases, identify phases that are memory-bandwidth limited, and discuss alternatives for phases that may be sensitive to the trend to employ lower precisions for performance.

Boukaram, Wajih↗

The Reverse Cuthill-McKee Algorithm in Distributed-Memory

Ordering vertices of a graph is key to minimize fill-in and data structure size in sparse direct solvers, maximize locality in iterative solvers, and improve performance in graph algorithms. Except for naturally parallelizable ordering methods such as nested dissection, many important ordering methods have not been efficiently mapped to distributed-memory architectures. In this paper, we present the first-ever distributed-memory implementation of the reverse Cuthill-McKee (RCM) algorithm for reducing the profile of a sparse matrix. Our parallelization uses a two-dimensional sparse matrix decomposition. We achieve high performance by decomposing the problem into a small number of primitives and utilizing optimized implementations of these primitives. Our implementation shows strong scaling up to 1024 cores for smaller matrices and up to 4096 cores for larger matrices.

97 MATHEMATICS AND COMPUTING↗

An Algebraic Multilevel Preconditioner with Low-Rank Corrections for Sparse Symmetric Matrices

In this work, we describe a multilevel preconditioning technique for solving sparse symmetric linear systems of equations. This “Multilevel Schur Low-Rank” (MSLR) preconditioner first builds a tree structure $\mathcal{T}$ based on a hierarchical decomposition of the matrix and then computes an approximate inverse of the original matrix level by level. Unlike classical direct solvers, the construction of the MSLR preconditioner follows a top-down traversal of $\mathcal{T}$ and exploits a low-rank property that is satisfied by the difference between the inverses of the local Schur complements and specific blocks of the original matrix. A few steps of the generalized Lanczos tridiagonalization procedure are applied to capture most of this difference. Numerical results are reported to illustrate the efficiency and robustness of the MSLR preconditioner with both two- and three-dimensional discretized PDE problems and with publicly available test problems.

97 MATHEMATICS AND COMPUTING↗

symPACK v2.0

SymPACK is a parallel direct solver for sparse symmetric systems of linear equations. SymPACK is implemented for large- scale distributed-memory platforms. It employs several innovative ideas, including the use of an asynchronous task paradigm, one-sided communication, and dynamic scheduling.

Ng, EsmondG↗

Tacho

SAND2022-7470 O Tacho is a performance portable sparse direct solver for use with various sparse matrix problems which typically arise from finite element methods. The code relies on the Kokkos parallel programming model porting to both CPUs and GPUs. It is also part of the Trilinos library. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Kim, Kyungjoo↗

Acceleration of the IMplicit–EXplicit nonhydrostatic unified model of the atmosphere on manycore processors

We present the acceleration of an IMplicit-EXplicit (IMEX) nonhydrostatic atmospheric model on manycore processors such as graphic processing units (GPUs) and Intel's Many Integrated Core (MIC) architecture. IMEX time integration methods sidestep the constraint imposed by the Courant-Friedrichs-Lewy condition on explicit methods through corrective implicit solves within each time step. In this work, we implement and evaluate the performance of IMEX on manycore processors relative to explicit methods. Using 3D-IMEX at Courant number C = 15, we obtained a speedup of about 4x relative to an explicit time stepping method run with the maximum allowable C = 1. Moreover, the unconditional stability of IMEX with respect to the fast waves means the speedup can increase significantly with the Courant number as long as the accuracy of the resulting solution is acceptable. We show a speedup of 100x at C = 150 using 1D-IMEX to demonstrate this point. Several improvements on the IMEX procedure were necessary in order to outperform our results with explicit methods: (a) reducing the number of degrees of freedom of the IMEX formulation by forming the Schur complement, (b) formulating a horizontally explicit vertically implicit 1D-IMEX scheme that has a lower workload and better scalability than 3D-IMEX, (c) using high-order polynomial preconditioners to reduce the condition number of the resulting system, and (d) using a direct solver for the 1D-IMEX method by performing and storing LU factorizations once to obtain a constant cost for any Courant number. Without all of these improvements, explicit time integration methods turned out to be difficult to beat. We discuss in detail the IMEX infrastructure required for formulating and implementing efficient methods on manycore processors. Several parametric studies are conducted to demonstrate the gain from each of the abovementioned improvements. Lastly, we validate our results with standard benchmark problems in numerical weather prediction and evaluate the performance and scalability of the IMEX method using up to 4192 GPUs and 16 Knights Landing processors.

97 MATHEMATICS AND COMPUTING↗

Verification of 2-D Detonation Shock Dynamics in conjunction with Los Alamos Lagrangian hydrocode

As the latest version of the fast-tube Detonation Shock Dynamics (DSD) solver is linked with the Los Alamos Lagrangian hydrocode, verification problems from a 2006 DSD report (LA-14277) have been duplicated with some of the verification criteria changed to more quantitative ones. The observed error convergence is as good as or better than reported in, quite possibly due to the careful treatment of floating point numbers to ensure that their precision level is maintained throughout the code. This report duplicates the three sample verification problems in LA-14277 using the Los Alamos ASC Lagrangian hydrocode (FLAG), official release of 3.2 Alpha6 with a few modifications. This version of FLAG is linked with the latest fast-tube Detonation Shock Dynamics (DSD) version beta 2 solver released in 2011 as part of the LanlDSD software product. New verification criteria are used for the arcwave problem where two specific locations are chosen for burn arrival time comparison. For this report FLAG's internal driver code prepares the distance function ({Psi}) and material ID fields from its hydro setup, instead of the stand-alone driver that is being utilized by the other LANL hydrocodes currently interfaced to LanlDSD. As it is implemented in version 3.2 Alpha6, the {Psi} and material ID fields and other parameters are passed from FLAG to the DSD solver directly, and the burn table is directly passed back to FLAG as part of the calling arguments. The burn-front arrival time 'exact' solutions, mentioned in the sequel for the rate-stick and 'arc-wave' problems, are computed using a pair of special-purpose Fortran codes provided by Aslam. In each case an ansatz for the form of the solution is made in which the radius from the detonator center point is used as the independent space coordinate. This leads to a simplified, problem-specific, 1D form of the governing equation. This equation is solved using 2nd-order spatial differencing and the forward Euler method on a very fine temporal and geometric mesh. The boundary conditions are handled exactly at the correct location, with second order accuracy. Care has been taken to ensure that this solution is fully converged.

36 MATERIALS SCIENCE↗

Improvements to CTF for Modeling of Boiling Water Reactor Geometry and Operating Conditions

This report documents work done to improve CTF modeling of boiling water reactor (BWR) designs during FY 2020. Several improvements were made to the geometry modeling capabilities, including adding support for modeling of inlet flow orifices, axially varying large water rods, channel boxes with large corner round radii, and quarter symmetry models. Work was also done to improve on the numerical stability and run-time performance of CTF in Virtual Environment for Reactor Applications (VERA) for core-scale BWR problems. An outer iteration loop was implemented that ensures mass and energy are conserved before moving on to the next time step in the solution. The ability to use a direct solver for the pressure matrix was also added, which reduces the matrix solution error and improves code stability and run-time performance. Improvements to the pressure balance iteration loop were needed to allow CTF to be used in a coupled VERA solution of a model of Hatch that was developed. Finally, various improvements have led to a 2×reduction in the total VERA walltime for the Hatch model.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

An implicit algorithm for the transonic full-potential equation in conservative form

A fast, implicit approximate factorization algorithm for the solution of the conservative full-potential equation for transonic flow in two and three dimensions is presented. Stability in supersonic regions is maintained by the use of an upwind evaluation of the density coefficient along all coordinate directions, providing an effective upwind difference of the streamwise terms for any orientation of the velocity vector and thereby enhancing the reliability of the algorithm. The algorithm is shown to provide rapid convergence for the computation of certain difficult two-dimensional test cases, including cases with fishtail shock patterns, demonstrating the reliability and efficiency of the procedure. Surface pressure coefficient distributions obtained by the present method are also found to be in good agreement with those computed by successive-line overrelaxation and a hybrid direct-solver/successive-line overrelaxation scheme, with significant reductions in CPU time required. A three-dimensional solution for a swept wing mounted between parallel walls is also presented which demonstrates the high convergence rate of the algorithm in three dimensions as well as two.

Holst, T.↗

Viscous compressible flow direct and inverse computation and illustrations

An algorithm for laminar and turbulent viscous compressible two dimensional flows is presented. For the application of precise boundary conditions over an arbitrary body surface, a body-fitted coordinate system is used in the physical plane. A thin-layer approximation of tne Navier-Stokes equations is introduced to keep the viscous terms relatively simple. The flow field computation is performed in the transformed plane. A factorized, implicit scheme is used to facilitate the computation. Sample calculations, for Couette flow, developing pipe flow, an isolated airflow, two dimensional compressor cascade flow, and segmental compressor blade design are presented. To a certain extent, the effective use of the direct solver depends on the user's skill in setting up the gridwork, the time step size and the choice of the artificial viscosity. The design feature of the algorithm, an iterative scheme to correct geometry for a specified surface pressure distribution, works well for subsonic flows. A more elaborate correction scheme is required in treating transonic flows where local shock waves may be involved.

Yang, T. T.↗

Aerothermal modeling program, phase 2

The main objective of the NASA sponsored Aerothermal Modeling Program, Phase 2--Element A, is to develop an improved numerical scheme for predicting combustor flow fields. This effort consists of the following three technical tasks. Task 1 involves the selection and evaluation of various candidate numerical techniques. Task 2 involves an in-depth evaluation of the selected numerical schemes. Task 3 involves the convection-diffusion scheme and the direct solver that will be incorporated in the NASA 3-D elliptic code (COM3S).

Karki, K. C.↗

Oscillatory flow with heat transfer in a square cavity

A computational study is presented for the flow inside an oscillatory cavity. The numerical scheme employs a semi-implicit, time-splitting method to integrate the two-dimensional full Navier-Stokes equations satisfying continuity to machine accuracy. The efficient use of direct solvers for the uncoupled momentum and pressure equations is demonstrated. The oscillatory cavity flow is studied considering the effects of heat transfer, Reynolds number, and oscillatory Stokes number.

Biringen, S.↗