使用Kokkos的低阶GPU加速有限元内核性能分析
Performance Analysis of Low-Order, GPU-accelerated Finite Element Kernels using Kokkos
- LMU Munich（慕尼黑大学）
- CERFACS（法国欧洲研究与教学中心流体力学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究分析低阶无矩阵有限元内核在多种GPU上的性能可移植性，发现内存映射是关键，并利用调优杠杆缩小性能差距。
AI中文摘要：
我们以地球物理模型中出现的矢量变系数偏微分算子为例，研究了低阶无矩阵有限元内核的性能可移植性。该内核使用Kokkos编写，在NVIDIA H100、AMD MI250X、AMD MI300A和Intel PVC Max 1550 GPU上进行了比较。由于其低阶以及减少算术运算的优化，该内核的算术强度较低，因此其性能取决于有限元组装如何映射到内存层次结构。这是一个即使在同一供应商系列内架构也不同的维度，导致不同的性能特征。我们考察了Kokkos的分层并行性和共享暂存内存（用于协调离散化的共享自由度）在每个设备上的表现。最后，我们展示了如何通过调整线程组大小、占用率与寄存器使用之间的平衡以及内核末尾的原子累加策略等调优杠杆来缩小可移植性差距。
英文摘要：
We study performance portability for low-order, matrix-free finite element kernels, using the example of a vectorial, variable-coefficient PDE operator originating in geophysical models. Written in Kokkos, the kernel is compared on NVIDIA H100, AMD MI250X, AMD MI300A and Intel PVC Max 1550 GPUs. Owing to its low order and to optimizations that reduce the arithmetic, the kernel has a low arithmetic intensity, so that its performance is determined by how the finite element assembly is mapped onto the memory hierarchy. This is a dimension in which the architectures differ even within one vendor family, causing different performance characteristics. We examine how Kokkos' hierarchical parallelism and shared scratch memory, which are used for the shared degrees of freedom of the conforming discretization, behave on each device. Finally, we show how portability gaps can be narrowed with tuning levers such as the size of the thread groups, the balance between occupancy and register use, and the atomic accumulation strategy at the end of the kernel.
