arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.01597cs.DCphysics.comp-phphysics.flu-dyn
收藏

使用CUDA感知的MPI进行高阶模板计算的可扩展通信

Scalable communication for high-order stencil computations using CUDA-aware MPI

Johannes Pekkilä, Miikka S. Väisälä, Maarit J. Käpylä, Matthias Rheinhardt, Oskar Lappi

摘要:

高性能计算中的现代计算节点提供了巨大的并行性和处理能力。然而,由于已经观察到算术性能相对于存储器和网络带宽以更快的速率增加,因此优化数据移动对于在许多通信繁重的应用中实现强大的可扩展性变得至关重要。随着图形处理单元的引入,这种性能差距进一步加剧,图形处理单元可以通过多种因素提供比中央处理单元更高的数据并行任务吞吐量。在这项工作中,我们探讨了迭代模板循环的计算方面,并实现了一个通用的通信方案,使用CUDA感知MPI,我们用它来加速基于高阶有限差分和三阶龙格库塔积分的磁流体力学模拟。我们特别关注提高工作负载的节点内局部性。我们的GPU实现可在$50\%$上从一个$64$设备强烈扩展--基于理论性能模型的预期效率的$87\%$。与多核CPU求解器相比,我们的实现在$16$节点上的计算约束基准测试中表现出$20$-$60\times$加速和$9$ - $12\times$提高的能量效率。

英文摘要:

Modern compute nodes in high-performance computing provide a tremendous level of parallelism and processing power. However, as arithmetic performance has been observed to increase at a faster rate relative to memory and network bandwidths, optimizing data movement has become critical for achieving strong scaling in many communication-heavy applications. This performance gap has been further accentuated with the introduction of graphics processing units, which can provide by multiple factors higher throughput in data-parallel tasks than central processing units. In this work, we explore the computational aspects of iterative stencil loops and implement a generic communication scheme using CUDA-aware MPI, which we use to accelerate magnetohydrodynamics simulations based on high-order finite differences and third-order Runge-Kutta integration. We put particular focus on improving intra-node locality of workloads. Our GPU implementation scales strongly from one to $64$ devices at $50\%$--$87\%$ of the expected efficiency based on a theoretical performance model. Compared with a multi-core CPU solver, our implementation exhibits $20$--$60\times$ speedup and $9$--$12\times$ improved energy efficiency in compute-bound benchmarks on $16$ nodes.

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。

↑