Problem
Research questions and friction points this paper is trying to address.
agentic science
resource discovery
hierarchical architecture
workload scheduling
heterogeneous systems
Hierarchical Server Architecture for Agentic Science
🤖 AI Summary
This work addresses the challenge of automated resource discovery and scheduling across heterogeneous, multi-institutional computing environments—including cloud, edge, and high-performance computing (HPC) infrastructures—in agent-based systems. The authors propose a hierarchical dynamic agent architecture in which secretary agents concurrently and asynchronously perform resource probing, negotiation, and task dispatching. By integrating an asynchronous negotiation protocol, agent-driven resource categorization, and a dynamic scheduling algorithm, the framework enables highly scalable, cross-infrastructure automation. Evaluated on a testbed comprising 51 real and simulated resource providers, the approach achieved a negotiation accuracy of 87.71% over 19,973 negotiation rounds and 6,952 task selections, with task selection costs comparable to conventional strategies, thereby significantly enhancing scheduling efficiency and adaptability.
📝 Abstract
Agentic science is transforming the landscape of computational work, extending to scientific pipelines and workload managers. The workloads require specialized hardware within and across institutions. If assessing workload needs against environments is required for scheduling, automated discovery of resources is an essential step. In this paper, we present a hierarchical, dynamic architecture and software to discover resources across diverse cloud, edge, and HPC systems. The design enables concurrent, asynchronous negotiation, selection, and dispatch of requests for work using secretary agents. The agents probe and discover 51 real and simulated providers across 7 categories. We perform 19,973 negotiation and 6,952 selection simulations to assess reliability of decisions, demonstrating high (87.71\%) negotiation accuracy and selection costs comparable to more traditional strategies. Designed for extensibility and currently supporting the Genesis Mission, this architecture exemplifies the importance of careful coordination between agents, discovery tools, and infrastructure for agentic science.
Innovation
Methods, ideas, or system contributions that make the work stand out.
hierarchical architecture
agentic science
resource discovery
secretary agents
asynchronous negotiation
🔎 Similar Papers
