An Exploration of OpenCL for a Numerical Relativity Application

TL;DR

This paper investigates the use of OpenCL for a Numerical Relativity application, demonstrating significant performance gains on GPUs over CPUs and showing that a single GPU can rival small CPU clusters.

Contribution

It is the first comprehensive exploration of OpenCL for a Numerical Relativity PDE solver, highlighting portability and performance benefits across hardware platforms.

Findings

Order-of-magnitude performance improvements with GPUs

Single high-end GPU matches small CPU cluster performance

OpenCL enables portable and efficient scientific computing

Abstract

Currently there is considerable interest in making use of many-core processor architectures, such as Nvidia and AMD graphics processing units (GPUs) for scientific computing. In this work we explore the use of the Open Computing Language (OpenCL) for a typical Numerical Relativity application: a time-domain Teukolsky equation solver (a linear, hyperbolic, partial differential equation solver using finite-differencing). OpenCL is the only vendor-agnostic and multi-platform parallel computing framework that has been adopted by all major processor vendors. Therefore, it allows us to write portable source-code and run it on a wide variety of compute hardware and perform meaningful comparisons. The outcome of our experimentation suggests that it is relatively straightforward to obtain order-of-magnitude gains in overall application performance by making use of many-core GPUs over multi-core CPUs and this fact is largely independent of the specific hardware architecture and vendor. We also observe that a single high-end GPU can match the performance of a small-sized, message-passing based CPU cluster.