探索RISC-V长向量能力:地球科学案例研究
Exploring RISC-V Long Vector Capabilities: A Case Study in Earth Sciences
AI总结:
本文研究地球科学代码SeisSol和MiniFALL3D在RISC-V长向量CPU上的性能,通过批处理GEMM和源码修改优化向量化,证明RISC-V向量架构的潜力且修改对其他架构也有益。
摘要:
本文研究了地球科学代码(具体为SeisSol和MiniFALL3D)在配备长向量处理单元的RISC-V CPU上的性能。研究重点在于优化这些应用以提高计算效率,同时保持可移植性。对于SeisSol,我们探索了批处理GEMM实现,通过利用指令级并行性来提升性能。MiniFALL3D的优化涉及通过修改源代码来改善向量化,例如将函数替换为子程序并展平多维数组。向量化过程始终留给编译器,以确保代码可移植性。研究使用软件模拟器和名为EPAC的RISC-V向量架构硬件原型进行。在两个应用上评估了不同HPC平台的性能,包括EPAC(基于RISC-V)、MareNostrum 4(由Sapphire Rapids CPU驱动)和NEC SX-Aurora Tsubasa加速器。我们旨在为将地球科学代码适配现代高性能计算系统提供见解,同时展示RISC-V向量架构的潜力。最终,所有为在RISC-V长向量架构上提升性能所做的修改,在其他具有不同向量能力的HPC架构上也显示出益处。这凸显了在依赖编译器强大的自动向量化能力的同时,保持代码可移植性的重要性。
英文摘要:
This paper investigates the performance of Earth Sciences codes, specifically SeisSol and MiniFALL3D, on a RISC-V-based CPU featuring a long vector processing unit. The study focuses on optimizing these applications for improved computational efficiency while maintaining portability. For SeisSol, we explore batched GEMM implementations to enhance performance by leveraging instruction-level parallelism. MiniFALL3D's optimization involves improving vectorization by modifying the source code, such as replacing functions with subroutines and flattening multidimensional arrays. The vectorization process is always left to the compiler to ensure code portability. The study is conducted using both a software emulator and a hardware prototype of the RISC-V vector architecture called EPAC. The performance of both applications is evaluated across different HPC platforms, including EPAC (based on RISC-V), MareNostrum 4 (powered by Sapphire Rapids CPUs), and the NEC SX-Aurora Tsubasa accelerator. We aim to provide insights into adapting Earth Sciences codes for modern high-performance computing systems while demonstrating the potential of RISC-V vector architectures. Ultimately, all modifications made to improve performance on the RISC-V long vector architecture are shown to be beneficial on other HPC architectures with different vector capabilities. This highlights the importance of maintaining code portability while relying on the compiler's powerful auto-vectorization capabilities.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。