🤖 AI Summary
This work addresses the limitations of current high-performance computing (HPC) systems, which are predominantly optimized for batch processing and thus struggle to support interactive operations and real-time responses required by urgent tasks—constraints that hinder their applicability in simulation, data analysis, and machine learning. The paper presents the first systematic examination of the emerging convergence between interactivity and urgency in HPC, reviewing key techniques including interactive scheduling strategies, priority mechanisms for urgent tasks, human-in-the-loop frameworks, and adaptive resource management. Building on this synthesis, it proposes a pathway for deeply integrating these approaches into existing HPC ecosystems, identifies critical challenges, and outlines a roadmap toward next-generation systems characterized by high responsiveness and enhanced user engagement, thereby advancing HPC toward a real-time interactive paradigm.
📝 Abstract
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.