arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.09511cs.DCcs.PF
收藏

Morpheus 释放:新兴架构上的快速跨平台 SpMV

Morpheus unleashed: Fast cross-platform SpMV on emerging architectures

Christodoulos Stylianou, Mark Klaisoongnoen, Ricardo Jesus, Nick Brown, Michele Weiland

AI总结:

本文研究 COO、CSR、DIA 等稀疏矩阵存储格式在 AArch64 CPU 和 FPGA 上的适用性,提出硬件特定优化并集成到 Morpheus 库中,通过 HPCG 基准测试验证跨平台 SpMV 性能。

摘要:

稀疏矩阵和线性代数是科学模拟的核心。多年来,已经开发了 70 多种稀疏矩阵存储格式,面向广泛的硬件架构和矩阵类型,每种格式都利用了架构的特定优势或矩阵的特定稀疏模式。在这项工作中,我们探讨了 COO、CSR 和 DIA 等存储格式对 AArch64 CPU 和 FPGA 等新兴架构的适用性。此外,我们详细介绍了针对这些目标的硬件特定优化,并评估了每项贡献被集成到 Morpheus 中的潜力。Morpheus 是一个现代库,目前提供跨 x86 CPU 和 NVIDIA/AMD GPU 的稀疏矩阵抽象。最后,我们通过将启用 Morpheus 的 HPCG 基准测试与供应商优化实现进行性能比较来验证我们的工作。

英文摘要:

Sparse matrices and linear algebra are at the heart of scientific simulations. Over the years, more than 70 sparse matrix storage formats have been developed, targeting a wide range of hardware architectures and matrix types, each of which exploit the particular strengths of an architecture, or the specific sparsity patterns of the matrices. In this work, we explore the suitability of storage formats such as COO, CSR and DIA for emerging architectures such as AArch64 CPUs and FPGAs. In addition, we detail hardware-specific optimisations to these targets and evaluate the potential of each contribution to be integrated into Morpheus, a modern library that provides an abstraction of sparse matrices (currently) across x86 CPUs and NVIDIA/AMD GPUs. Finally, we validate our work by comparing the performance of the Morpheus-enabled HPCG benchmark against vendor-optimised implementations.

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。

↑