Parallel Algorithms for Successive Convolution

TL;DR

This paper develops parallel, matrix-free algorithms for PDE discretizations involving integral operators, demonstrating high efficiency, scalability, and stability in large-scale computing environments.

Contribution

It introduces a domain decomposition and hybrid MPI/Kokkos implementation for these PDE methods, enabling scalable performance on distributed and shared memory systems.

Findings

Achieves over 10^8 DOF/node/s update rate.

Demonstrates scalability across various PDE problems.

Ensures stability and efficiency in large-scale computations.

Abstract

In this work, we consider alternative discretizations for PDEs which use expansions involving integral operators to approximate spatial derivatives. These constructions use explicit information within the integral terms, but treat boundary data implicitly, which contributes to the overall speed of the method. This approach is provably unconditionally stable for linear problems and stability has been demonstrated experimentally for nonlinear problems. Additionally, it is matrix-free in the sense that it is not necessary to invert linear systems and iteration is not required for nonlinear terms. Moreover, the scheme employs a fast summation algorithm that yields a method with a computational complexity of $\mathcal{O}(N)$, where $N$ is the number of mesh points along a direction. While much work has been done to explore the theory behind these methods, their practicality in large scale computing environments is a largely unexplored topic. In this work, we explore the performance of these methods by developing a domain decomposition algorithm suitable for distributed memory systems along with shared memory algorithms. As a first pass, we derive an artificial CFL condition that enforces a nearest-neighbor communication pattern and briefly discuss possible generalizations. We also analyze several approaches for implementing the parallel algorithms by optimizing predominant loop structures and maximizing data reuse. Using a hybrid design that employs MPI and Kokkos for the distributed and shared memory components of the algorithms, respectively, we show that our methods are efficient and can sustain an update rate $> 1\times10^8$ DOF/node/s. We provide results that demonstrate the scalability and versatility of our algorithms using several different PDE test problems, including a nonlinear example, which employs an adaptive time-stepping rule.