Comparing Performance and Portability between CUDA and SYCL for Protein Database Search on NVIDIA, AMD, and Intel GPUs

TL;DR

This study compares CUDA and SYCL for protein database search across NVIDIA, AMD, and Intel GPUs, showing SYCL's superior portability and comparable performance on NVIDIA devices.

Contribution

It demonstrates that SYCL offers better portability across diverse GPU architectures while maintaining similar performance levels to CUDA on NVIDIA GPUs.

Findings

SYCL achieves remarkable portability to AMD and Intel GPUs.

Performance on NVIDIA GPUs is similar between CUDA and SYCL.

SYCL outperforms in architectural efficiency on most tested devices.

Abstract

The heterogeneous computing paradigm has led to the need for portable and efficient programming solutions that can leverage the capabilities of various hardware devices, such as NVIDIA, Intel, and AMD GPUs. This study evaluates the portability and performance of the SYCL and CUDA languages for one fundamental bioinformatics application (Smith-Waterman protein database search) across different GPU architectures, considering single and multi-GPU configurations from different vendors. The experimental work showed that, while both CUDA and SYCL versions achieve similar performance on NVIDIA devices, the latter demonstrated remarkable code portability to other GPU architectures, such as AMD and Intel. Furthermore, the architectural efficiency rates achieved on these devices were superior in 3 of the 4 cases tested. This brief study highlights the potential of SYCL as a viable solution for achieving both performance and portability in the heterogeneous computing ecosystem.