Euro-Par 2022: Parallel Processing Workshops - Euro-Par 2022 International Workshops, Glasgow, UK, August 22-26, 2022, Revised Selected Papers
Euro-Par 2022: Parallel Processing Workshops - Euro-Par 2022 International Workshops, Glasgow, UK, August 22-26, 2022, Revised Selected Papers
复制标题
Euro-Par 2022：并行处理研讨会 - Euro-Par 2022 国际研讨会，英国格拉斯哥，2022 年 8 月 22-26 日，修订后的精选论文
DOI：
10.1007/978-3-031-31209-0_4
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Brown N
中科院分区：
文献类型：
--
作者：
Brown N
Cerebras Wafer Scale Engine (WSE) 是一款将数十万个 AI 核心组合到单个芯片上的加速器。虽然这项技术是为机器学习工作负载而设计的，但大量可用的原始计算意味着它也是加速传统 HPC 计算代码的一个非常有趣的潜在目标。其中许多算法都是基于模板的，其中更新操作涉及相邻元素的贡献，在本文中，与 CPU 和 GPU 相比，我们从该技术的早期采用者的角度探讨了该技术对此类代码的适用性。在 Cerebras CS-1 上运行，我们探索了性能并描述了程序员目前表达他们算法的方式。我们证明，虽然在向用户公开编程接口方面仍有工作要做，但 WSE 的性能令人印象深刻，因为在我们的实验中，它的性能比四个 V100 GPU 快两倍半，比两个 Intel Xeon Platinum CPU 快约 114 倍。因此，这项技术在加速未来百亿亿次超级计算机上的 HPC 代码方面具有巨大的潜力。
The Cerebras Wafer Scale Engine (WSE) is an accelerator that combines hundreds of thousands of AI-cores onto a single chip. Whilst this technology has been designed for machine learning workloads, the significant amount of available raw compute means that it is also a very interesting potential target for accelerating traditional HPC computational codes. Many of these algorithms are stencil-based, where update operations involve contributions from neighbouring elements, and in this paper we explore the suitability of this technology for such codes from the perspective of an early adopter of the technology, compared to CPUs and GPUs. Running on a Cerebras CS-1 we explore the performance and describe in which programmers at the moment express their algorithms. We demonstrate that, whilst there is still work to be done around exposing the programming interface to users, performance of the WSE is impressive as it out performs four V100 GPUs by two and a half times and two Intel Xeon Platinum CPUs by around 114 times in our experiments. There is significant potential therefore for this technology to play an important role in accelerating HPC codes on future exascale supercomputers.