Exploring the Versal AI engines for accelerating stencil-based atmospheric advection simulation

TL;DR

This paper investigates the use of AMD's Versal AI engines to accelerate atmospheric advection simulations, demonstrating potential performance improvements and exploring optimal implementation strategies for stencil-based HPC workloads.

Contribution

It provides an initial exploration of porting a stencil-based atmospheric advection scheme onto Versal AI engines, highlighting performance benefits and implementation considerations.

Findings

AI engines can double performance over traditional FPGA configurations.

Performance is limited by channel bandwidth between fabric and AI engines.

Versal ACAP shows promise for accelerating HPC simulations.

Abstract

AMD Xilinx's new Versal Adaptive Compute Acceleration Platform (ACAP) is an FPGA architecture combining reconfigurable fabric with other on-chip hardened compute resources. AI engines are one of these and, by operating in a highly vectorized manner, they provide significant raw compute that is potentially beneficial for a range of workloads including HPC simulation. However, this technology is still early-on, and as yet unproven for accelerating HPC codes, with a lack of benchmarking and best practice. This paper presents an experience report, exploring porting of the Piacsek and Williams (PW) advection scheme onto the Versal ACAP, using the chip's AI engines to accelerate the compute. A stencil-based algorithm, advection is commonplace in atmospheric modelling, including several Met Office codes who initially developed this scheme. Using this algorithm as a vehicle, we explore optimal approaches for structuring AI engine compute kernels and how best to interface the AI engines with programmable logic. Evaluating performance using a VCK5000 against non-AI engine FPGA configurations on the VCK5000 and Alveo U280, as well as a 24-core Xeon Platinum Cascade Lake CPU and Nvidia V100 GPU, we found that whilst the number of channels between the fabric and AI engines are a limitation, by leveraging the ACAP we can double performance compared to an Alveo U280.