Real-Time AI Service Economy: A Framework for Agentic Computing Across the Continuum

Researchers propose a framework for decentralized resource allocation in real-time AI services across device-edge-cloud infrastructure. The study shows that dependency graph topology determines whether price-based allocation can work at scale, with hierarchical structures enabling stable pricing while complex dependencies cause instability.

• →Dependency graph topology is the primary factor determining whether decentralized AI resource allocation can scale reliably.
• →Hierarchical service dependencies (tree-like structures) enable stable price convergence and efficient resource allocation.
• →Complex cross-cutting dependencies between pipeline stages cause price oscillations and degraded allocation quality.
• →A hybrid architecture using cross-domain integrators can reduce price volatility by 70-75% without sacrificing throughput.
• →Decentralized coordination can match centralized allocation quality under truthful bidding mechanisms.