Showing 1–4 of 4 results for author: Kuckuk, S

p-adaptive discontinuous Galerkin method for the shallow water equations on heterogeneous computing architectures

Authors: Sara Faghih-Naini, Vadym Aizinger, Sebastian Kuckuk, Richard Angersbach, Harald Köstler

Abstract: Heterogeneous computing and exploiting integrated CPU-GPU architectures has become a clear current trend since the flattening of Moore's Law. In this work, we propose a numerical and algorithmic re-design of a p-adaptive quadrature-free discontinuous Galerkin method (DG) for the shallow water equations (SWE). Our new approach separates the computations of the non-adaptive (lower-order) and adaptiv… ▽ More Heterogeneous computing and exploiting integrated CPU-GPU architectures has become a clear current trend since the flattening of Moore's Law. In this work, we propose a numerical and algorithmic re-design of a p-adaptive quadrature-free discontinuous Galerkin method (DG) for the shallow water equations (SWE). Our new approach separates the computations of the non-adaptive (lower-order) and adaptive (higher-order) parts of the discretization form each other. Thereby, we can overlap computations of the lower-order and the higher-order DG solution components. Furthermore, we investigate execution times of main computational kernels and use automatic code generation to optimize their distribution between the CPU and GPU. Several setups, including a prototype of a tsunami simulation in a tide-driven flow scenario, are investigated, and the results show that significant performance improvements can be achieved in suitable setups. △ Less

Submitted 19 November, 2023; originally announced November 2023.

Optimizing Geometric Multigrid Methods with Evolutionary Computation

Authors: Jonas Schmitt, Sebastian Kuckuk, Harald Köstler

Abstract: For many linear and nonlinear systems that arise from the discretization of partial differential equations the construction of an efficient multigrid solver is a challenging task. Here we present a novel approach for the optimization of geometric multigrid methods that is based on evolutionary computation, a generic program optimization technique inspired by the principle of natural evolution. A m… ▽ More For many linear and nonlinear systems that arise from the discretization of partial differential equations the construction of an efficient multigrid solver is a challenging task. Here we present a novel approach for the optimization of geometric multigrid methods that is based on evolutionary computation, a generic program optimization technique inspired by the principle of natural evolution. A multigrid solver is represented as a tree of mathematical expressions which we generate based on a tailored grammar. The quality of each solver is evaluated in terms of convergence and compute performance using automated local Fourier analysis (LFA) and roofline performance modeling, respectively. Based on these objectives a multi-objective optimization is performed using strongly typed genetic programming with a non-dominated sorting based selection. To evaluate the model-based prediction and to target concrete applications, scalable implementations of an evolved solver can be automatically generated with the ExaStencils framework. We demonstrate our approach by constructing multigrid solvers for the steady-state heat equation with constant and variable coefficients that consistently perform better than common V- and W-cycles. △ Less

Submitted 8 October, 2019; v1 submitted 7 October, 2019; originally announced October 2019.

Towards whole program generation of quadrature-free discontinuous Galerkin methods for the shallow water equations

Authors: Sara Faghih-Naini, Sebastian Kuckuk, Vadym Aizinger, Daniel Zint, Roberto Grosso, Harald Köstler

Abstract: The shallow water equations (SWE) are a commonly used model to study tsunamis, tides, and coastal ocean circulation. However, there exist various approaches to discretize and solve them efficiently. Which of them is best for a certain scenario is often not known and, in addition, depends heavily on the used HPC platform. From a simulation software perspective, this places a premium on the ability… ▽ More The shallow water equations (SWE) are a commonly used model to study tsunamis, tides, and coastal ocean circulation. However, there exist various approaches to discretize and solve them efficiently. Which of them is best for a certain scenario is often not known and, in addition, depends heavily on the used HPC platform. From a simulation software perspective, this places a premium on the ability to adapt easily to different numerical methods and hardware architectures. One solution to this problem is to apply code generation techniques and to express methods and specific hardware-dependent implementations on different levels of abstraction. This allows for a separation of concerns and makes it possible, e.g., to exchange the discretization scheme without having to rewrite all low-level optimized routines manually. In this paper, we show how code for an advanced quadrature-free discontinuous Galerkin (DG) discretized shallow water equation solver can be generated. Here, we follow the multi-layered approach from the ExaStencils project that starts from the continuous problem formulation, moves to the discrete scheme, spells out the numerical algorithms, and, finally, maps to a representation that can be transformed to a distributed memory parallel implementation by our in-house Scala-based source-to-source compiler. Our contributions include: A new quadrature-free discontinuous Galerkin formulation, an extension of the class of supported computational grids, and an extension of our toolchain allowing to evaluate discrete integrals stemming from the DG discretization implemented in Python. As first results we present the whole toolchain and also demonstrate the convergence of our method for higher order DG discretizations. △ Less

Submitted 18 April, 2019; originally announced April 2019.

A Scala Prototype to Generate Multigrid Solver Implementations for Different Problems and Target Multi-Core Platforms

Authors: Harald Koestler, Christian Schmitt, Sebastian Kuckuk, Frank Hannig, Juergen Teich, Ulrich Ruede

Abstract: Many problems in computational science and engineering involve partial differential equations and thus require the numerical solution of large, sparse (non)linear systems of equations. Multigrid is known to be one of the most efficient methods for this purpose. However, the concrete multigrid algorithm and its implementation highly depend on the underlying problem and hardware. Therefore, changes… ▽ More Many problems in computational science and engineering involve partial differential equations and thus require the numerical solution of large, sparse (non)linear systems of equations. Multigrid is known to be one of the most efficient methods for this purpose. However, the concrete multigrid algorithm and its implementation highly depend on the underlying problem and hardware. Therefore, changes in the code or many different variants are necessary to cover all relevant cases. In this article we provide a prototype implementation in Scala for a framework that allows abstract descriptions of PDEs, their discretization, and their numerical solution via multigrid algorithms. From these, one is able to generate data structures and implementations of multigrid components required to solve elliptic PDEs on structured grids. Two different test problems showcase our proposed automatic generation of multigrid solvers for both CPU and GPU target platforms. △ Less

Submitted 20 June, 2014; originally announced June 2014.