Field-Programmable Gate Arrays and Quantum Monte Carlo: Power Efficient Co-processing for Scalable High-Performance Computing

TL;DR

This paper explores the use of FPGAs to enhance power efficiency in high-performance scientific computing, specifically for Variational Monte Carlo, aiming to support exascale computing within sustainable energy limits.

Contribution

It demonstrates the potential of FPGA co-processing to improve performance per power unit in scientific applications like Variational Monte Carlo, advancing exascale computing feasibility.

Findings

FPGA co-processing improves power efficiency for Variational Monte Carlo.

FPGAs can complement multicore systems to achieve scalable high-performance computing.

Potential for FPGAs to be integral in future exascale platforms.

Abstract

Massively parallel architectures offer the potential to significantly accelerate an application relative to their serial counterparts. However, not all applications exhibit an adequate level of data and/or task parallelism to exploit such platforms. Furthermore, the power consumption associated with these forms of computation renders "scaling out" for exascale levels of performance incompatible with modern sustainable energy policies. In this work, we investigate the potential for field-programmable gate arrays (FPGAs) to feature in future exascale platforms, and their capacity to improve performance per unit power measurements for the purposes of scientific computing. We have focussed our efforts on Variational Monte Carlo, and report on the benefits of co-processing with an FPGA relative to a purely multicore system.