标题： 使用混合精度跨越内存墙 -- 在百亿亿次机器上的HPG-MxP 显示英文标题

标题： Scaling the memory wall using mixed-precision -- HPG-MxP on an exascale machine

摘要： 混合精度算法被提出作为一种方式，使科学计算能够从最近高性能计算（HPC）平台上看到的人工智能（AI）的一些优势中受益。 一些以密集矩阵运算为主的应用程序通过利用低精度格式如FP16已经看到了显著的加速。 然而，大多数科学模拟应用程序受内存带宽限制。 除了初步研究外，在给定的HPC系统上使用混合精度算法的实际收益仍然 largely 不清楚。 高性能GMRES混合精度（HPG-MxP）基准已被提出，以衡量HPC系统在基于稀疏矩阵的混合精度应用程序中的有用性能。 在这项工作中，我们为exascale系统提出了HPG-MxP基准的高效优化实现，并描述了我们的算法增强。 我们首次展示了在现代基于GPU的超级计算机上结合双精度和单精度使用的1.6倍加速。 显示英文摘要

摘要： Mixed-precision algorithms have been proposed as a way for scientific computing to benefit from some of the gains seen for artificial intelligence (AI) on recent high performance computing (HPC) platforms. A few applications dominated by dense matrix operations have seen substantial speedups by utilizing low precision formats such as FP16. However, a majority of scientific simulation applications are memory bandwidth limited. Beyond preliminary studies, the practical gain from using mixed-precision algorithms on a given HPC system is largely unclear. The High Performance GMRES Mixed Precision (HPG-MxP) benchmark has been proposed to measure the useful performance of a HPC system on sparse matrix-based mixed-precision applications. In this work, we present a highly optimized implementation of the HPG-MxP benchmark for an exascale system and describe our algorithm enhancements. We show for the first time a speedup of 1.6x using a combination of double- and single-precision on modern GPU-based supercomputers.