Efficient Implementation of Nonlinear Compact Schemes on Massively Parallel Platforms

  title={Efficient Implementation of Nonlinear Compact Schemes on Massively Parallel Platforms},
  author={Debojyoti Ghosh and Emil M. Constantinescu and Jed Brown},
  journal={SIAM J. Sci. Comput.},
Weighted nonlinear compact schemes are ideal for simulating compressible, turbulent flows because of their nonoscillatory nature and high spectral resolution. However, they require the solution to banded systems of equations at each time-integration step or stage. We focus on tridiagonal compact schemes in this paper. We propose an efficient implementation of such schemes on massively parallel computing platforms through an iterative substructuring algorithm to solve the tridiagonal system of… 

Nonlinear Compact Finite-Difference Schemes with Semi-Implicit Time Stepping

This work proposes a finite-difference method for atmospheric flows that uses a weighted compact scheme for spatial discretization and implicit-explicit additive Runge-Kutta methods for time integration.

Compact Schemes for Multiscale Flows with Cell-Centered Finite Difference Method

High-order compact interpolation schemes appropriate for multiscale flows are studied within a cell-centered finite difference method (CCFDM) framework where the robustness of high-order schemes on

Scalable Many-Core Algorithms for Tridiagonal Solvers

Performance comparisons with the state of the art, using both a large Cray EX system and a GPU cluster show the algorithmic tradeoffs required at increasing machine scale to achieve good performance, particularly considering the advent of exascale systems.

Well-Balanced Formulation of Gravitational Source Terms for Conservative Finite-Difference Atmospheric Flow Solvers

Numerical simulation of atmospheric flows requires high-resolution, nonoscillatory algorithms to accurately capture all length scales. In this paper, a conservative finite-difference algorithm is

Well-Balanced, Conservative Finite Difference Algorithm for Atmospheric Flows

The numerical simulation of meso-, convective-, and microscale atmospheric flows requires the solution of the Euler or the Navier–Stokes equations. Nonhydrostatic weather prediction algorithms often

Semi-Implicit Time Integration of Atmospheric Flows with Characteristic-Based Flux Partitioning

A characteristic-based flux partitioning for the semi-implicit time integration of atmospheric flows using high-order, conservative additive Runge--Kutta methods to integrate the acoustic and advective components of the hyperbolic flux in time implicitly with an unconditionally stable method.

GPU-accelerated DNS of compressible turbulent flows



A Higher-Order Compact Method in Space and Time Based on Parallel Implementation of the Thomas Algorithm

A novel method to parallelize high-order compact numerical algorithms for the solution of three-dimensional PDEs in a space?time domain that is driven by a communication and computation schedule instead of the usual “creative programming” approach is proposed.

Tera-Scalable Algorithms for Variable-Density Elliptic Hydrodynamics with Spectral Accuracy

This effort represents the first time that a high-order variable-density incompressible flow solver with species diffusion has demonstrated sustained performance in the TeraFLOPS range.

Practical Parallel Band Triangular System Solvers

A new algorithm for the fast solution of hnear recurrence systems, which is discussed in the form of band triangular linear systems, and variations on it which hold for certain special cases of practical interest are given.

A massively parallel multi-block hybrid compact-WENO scheme for compressible flows

Efficient Implementation of Weighted ENO Schemes

A new way of measuring the smoothness of a numerical solution is proposed, emulating the idea of minimizing the total variation of the approximation, which results in a fifth-order WENO scheme for the caser= 3, instead of the fourth-order with the original smoothness measurement by Liuet al.

An Efficient Parallel Algorithm for the Solution of a Tridiagonal Linear System of Equations

An efficient parallel algorithm is presented in which computation time grows as log 2, which can be used to solve recurrence relations of all orders.

Solving the compressible Navier-Stokes equations on up to 1.97 million cores and 4.1 trillion grid points

The use of hyperthreading is discussed, which significantly improves the parallel performance of the code of the Hybrid code, a finite-difference solver of the compressible Navier-Stokes equations on structured grids used for the direct numerical simulation of isotropic turbulence and its interaction with shock waves.

Implementation in ScaLAPACK of Divide-and-Conquer Algorithms forBanded and Tridiagonal Linear Systems

Comparison with existing dense-type methods shows that for areas of the problem parameter space with low bandwidth and/or high number of processors, the family of algorithms described here is superior.

Application of Compact-Reconstruction Weighted Essentially Nonoscillatory Schemes to Compressible Aerodynamic Flows

Compact-reconstruction weighted essentially nonoscillatory schemes have lower dissipation and dispersion errors as well as higher spectral resolution than weighted essentially nonoscillatory schemes