CuPBoP: CUDA for Parallelized and Broad-range Processors

TL;DR

CuPBoP enables CUDA programs to run on non-NVIDIA hardware without manual code modifications, achieving high coverage and performance across various architectures, thus broadening CUDA's applicability in heterogeneous systems.

Contribution

CuPBoP introduces a novel approach to execute CUDA on non-NVIDIA devices without manual source code modifications, surpassing existing frameworks in coverage and supporting multiple ISAs with competitive performance.

Findings

Achieves 69.6% coverage on Rodinia benchmark, higher than existing frameworks.

Supports multiple CPU ISAs like X86, RISC-V, AArch64 with high performance.

Comparable or superior performance to manually optimized OpenMP/MPI and CUDA programs.

Abstract

CUDA is one of the most popular choices for GPU programming, but it can only be executed on NVIDIA GPUs. Executing CUDA on non-NVIDIA devices not only benefits the hardware community, but also allows data-parallel computation in heterogeneous systems. To make CUDA programs portable, some researchers have proposed using source-to-source translators to translate CUDA to portable programming languages that can be executed on non-NVIDIA devices. However, most CUDA translators require additional manual modifications on the translated code, which imposes a heavy workload on developers. In this paper, CuPBoP is proposed to execute CUDA on non-NVIDIA devices without relying on any portable programming languages. Compared with existing work that executes CUDA on non-NVIDIA devices, CuPBoP does not require manual modification of the CUDA source code, but it still achieves the highest coverage (69.6%), much higher than existing frameworks (56.6%) on the Rodinia benchmark. In particular, for CPU backends, CuPBoP supports several ISAs (e.g., X86, RISC-V, AArch64) and has close or even higher performance compared with other projects. We also compare and analyze the performance among CuPBoP, manually optimized OpenMP/MPI programs, and CUDA programs on the latest Ampere architecture GPU, and show future directions for supporting CUDA programs on non-NVIDIA devices with high performance