以协处理器即服务实现 CMS 计算工作流的可移植加速
Portable acceleration of CMS computing workflows with coprocessors as a service
AI总结：
本文研究 CMS 实验中以 SONIC 即服务方式将机器学习推理任务卸载到 GPU 等协处理器，证明可显著加速算法并提升整体工作流吞吐量，同时保持可移植性。
摘要：
未来几十年，诸如 CERN LHC 上的 CMS 实验等大型科学实验的计算需求将急剧增长。为了补充未来中央处理器（CPU）上运行的软件性能提升，探索协处理器在数据处理中的使用具有巨大潜力和价值。协处理器是一类补充 CPU 的计算机处理器，通常由于架构设计选择而能改善某些功能的执行。我们探索了面向协处理器优化网络推理服务（SONIC）的方法，并研究了这种即服务方法在大规模数据处理中的部署。在研究中，我们采用 CMS 实验的一个数据处理工作流，在 CPU 上运行主工作流，同时将若干机器学习（ML）推理任务卸载到远程或本地协处理器上，具体为图形处理单元（GPU）。通过在 Google Cloud、Purdue Tier-2 计算中心以及两者组合环境中进行的实验，我们展示了这些 ML 算法在协处理器上的单独加速效果，以及整个工作流相应的吞吐量提升。该方法可以容易地推广到不同类型的协处理器，并部署在本地 CPU 上而不会降低吞吐量性能。我们强调，SONIC 方法能够实现较高的协处理器利用率，并使工作流具备在不同类型协处理器上运行的可移植性。
英文摘要：
Computing demands for large scientific experiments, such as the CMS experiment at the CERN LHC, will increase dramatically in the next decades. To complement the future performance increases of software running on central processing units (CPUs), explorations of coprocessor usage in data processing hold great potential and interest. Coprocessors are a class of computer processors that supplement CPUs, often improving the execution of certain functions due to architectural design choices. We explore the approach of Services for Optimized Network Inference on Coprocessors (SONIC) and study the deployment of this as-a-service approach in large-scale data processing. In the studies, we take a data processing workflow of the CMS experiment and run the main workflow on CPUs, while offloading several machine learning (ML) inference tasks onto either remote or local coprocessors, specifically graphics processing units (GPUs). With experiments performed at Google Cloud, the Purdue Tier-2 computing center, and combinations of the two, we demonstrate the acceleration of these ML algorithms individually on coprocessors and the corresponding throughput improvement for the entire workflow. This approach can be easily generalized to different types of coprocessors and deployed on local CPUs without decreasing the throughput performance. We emphasize that the SONIC approach enables high coprocessor usage and enables the portability to run workflows on different types of coprocessors.
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。
