This work argues for co-design, that facilities should treat agents as first-class principals where agents become facility-aware tenants, and outlines the resulting challenges and opportunities in compute, storage, agent memory, and safety.
Abstract
Coding agents have become real users of high-performance computing (HPC) systems, yet today's HPC abstractions, interfaces, and policies remain designed for human-driven workflows. In our measurement, users running coding agents are only 19.5% of the observed population, but account for 55.8% of job submissions, 29.1% of CPU core-hours, and 42.7% of GPU-hours. Agents are not simply faster humans. They issue commands at 20.8x the human rate, decompose work into fine-grained explore-modify-execute loops, and pursue open-ended goals through trial-and-error campaigns that continue through nights and weekends. These behaviors strain abstractions built for human timescales, surfacing as control-plane pressure on the scheduler, metadata-intensive I/O on bandwidth-provisioned filesystems, repeated rediscovery of what earlier sessions already learned, and new prompt-injection and policy-enforcement surfaces. Neither banning agents nor treating them as ordinary users is sustainable. We instead argue for co-design, that facilities should treat agents as first-class principals where agents become facility-aware tenants. We outline the resulting challenges and opportunities in compute, storage, agent memory, and safety.
RASER is presented, a user-space framework that enables seamless execution of agentic workflows on production HPC clusters by extending Slurm's internal primitives and provides resilience against preemption and failures while maintaining minimal checkpoint/restore overhead.
(English) High-performance computing (HPC) platforms are evolving towards increasingly complex architectures: many-core CPUs with multi-level NUMA hierarchies, heterogeneity with multiple classes of accelerators and higher-capacity interconnects. The increasing complexity and variety of resources in these machines make...
ASEND (Autonomous Scientific Computing Engine and Novel Discovery), an AI-powered agent system, operated through a command-line or a local web chat interface, that runs the agent on the researcher's own laptop, reaching Slurm clusters and a GPU workstation over multiplexed authenticated SSH, with site-specific executio...
J. P. Liu, Uthpala Herath, Andrew Petersen· 0 citations
HPC-AutoResearch is presented, a proof-of-concept system for the autonomous execution of compiled-code research workflows in HPC-like environments that divides this sub-pipeline into five phases—planning, environment setup, coding, compilation, and execution—localizing failures within each phase and enabling iterative...
T. Kotama, Shun-ichiro Hayashi, Daichi Mukunoki et al.· 1 citation
LLM4LLM is introduced, a deployment-aware closed-loop optimization framework that starts from a target inference script, extracts phase-aware optimization tasks, searches with an experience-guided episodic agent, and accepts patches through in-model validation.
Hui Zeng, Pengfei Yang, Yanxin Chen et al.· 0 citations
This paper presents a systematic refactoring approach using Claude agentic AI on real-world project-scheduling workloads in a high-performance computing (HPC) environment, where the agent identifies bottlenecks, implements targeted optimizations, and evaluates their effects, while the researcher retains final control.
H. Li, A. Pletzer, Yuan Tian et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.