Portability for GPU-accelerated molecular docking applications for cloud and HPC: can portable compiler directives provide performance across all platforms?

TL;DR

This paper investigates whether portable compiler directives can deliver consistent high performance for GPU-accelerated molecular docking applications across diverse hardware platforms, including NVIDIA, AMD, and Intel GPUs.

Contribution

It extends the MiniMDock miniapp to GPU offloading using OpenMP directives and compares its performance with CUDA and HIP implementations across multiple GPU vendors.

Findings

OpenMP directives can achieve comparable performance to CUDA and HIP on various GPUs.

Porting from device-specific code to directive-based code involves overcoming significant barriers.

The reverse-porting process is feasible but requires careful optimization.

Abstract

High-throughput structure-based screening of drug-like molecules has become a common tool in biomedical research. Recently, acceleration with graphics processing units (GPUs) has provided a large performance boost for molecular docking programs. Both cloud and high-performance computing (HPC) resources have been used for large screens with molecular docking programs; while NVIDIA GPUs have dominated cloud and HPC resources, new vendors such as AMD and Intel are now entering the field, creating the problem of software portability across different GPUs. Ideally, software productivity could be maximized with portable programming models that are able to maintain high performance across architectures. While in many cases compiler directives have been used as an easy way to offload parallel regions of a CPU-based program to a GPU accelerator, they may also be an attractive programming model for providing portability across different GPU vendors, in which case the porting process may proceed in the reverse direction: from low-level, architecture-specific code to higher-level directive-based abstractions. MiniMDock is a new mini-application (miniapp) designed to capture the essential computational kernels found in molecular docking calculations, such as are used in pharmaceutical drug discovery efforts, in order to test different solutions for porting across GPU architectures. Here we extend MiniMDock to GPU offloading with OpenMP directives, and compare to performance of kernels using CUDA, and HIP on both NVIDIA and AMD GPUs, as well as across different compilers, exploring performance bottlenecks. We document this reverse-porting process, from highly optimized device code to a higher-level version using directives, compare code structure, and describe barriers that were overcome in this effort.