When we think of how we use smartphones, e-commerce, collaboration platforms, LLMs, etc., most of our interactions with computers are interactive and often urgent. Similar trends of interactivity and urgency are coming to HPC, with applications from simulations to data analysis and machine learning requiring more parallel computational capability and more interactivity. This chapter overviews the progress made so far along with some vectors of what the path forward will bring for greater integration of interactive and urgent HPC policies, techniques, and technologies into our HPC ecosystems.

翻译：当我们思考智能手机、电子商务、协作平台、大语言模型（LLMs）等应用场景时，发现人与计算机的绝大多数交互都具有互动性与紧急性的特征。这种交互性与紧急性趋势正在向高性能计算（HPC）领域延伸——从模拟仿真到数据分析再到机器学习，各类应用不仅需要更强的并行计算能力，更要求更高的交互性。本章综述了当前已取得的研究进展，并探讨了将交互式与紧急HPC策略、方法和技术更深入地融入HPC生态系统的未来发展方向。