QCD on GPUs: cost effective supercomputing

TL;DR

This paper reviews how GPUs, with their high floating point power and low cost, are transforming lattice QCD calculations by requiring specialized algorithms to leverage their architecture effectively.

Contribution

It provides a comprehensive overview of the progress and challenges in deploying GPU-based computing for large-scale lattice QCD simulations.

Findings

GPUs offer a cost-effective platform for lattice QCD.

Unique algorithmic adaptations are necessary for GPU efficiency.

Progress has been made in large-scale GPU-based QCD calculations.

Abstract

The exponential growth of floating point power in graphics processing units (GPUs), together with their low cost, has given rise to an attractive platform upon which to deploy lattice QCD calculations. GPUs are essentially many (O(100)) core chips, that are programmed using a massively threaded environment, and so are representative of the future of high performance computing (HPC). The large ratio of raw floating point operations per second to memory bandwidth that is characteristic of GPUs necessitates that unique algorithmic design choices are made to harness their full potential. We review the progress to date in using GPUs for large scale calculations, and contrast GPUs against more traditional HPC architectures