Supporting Dynamic Agentic Workloads: How Data and Agents Interact

TL;DR

This paper introduces an Agent-Centric Data Fabric architecture designed to efficiently support dynamic, multi-modal workloads generated by large language model-powered multi-agent systems, addressing limitations of traditional static data management.

Contribution

It proposes a unified, adaptive data system architecture that leverages attention-guided retrieval, semantic micro-caching, and predictive prefetching to optimize agentic workloads.

Findings

Enhanced data access speed for agents

Reduced redundant queries and data movement

Improved efficiency in multi-agent interactions

Abstract

The rise of multi-agent systems powered by large language models (LLMs) and specialized reasoning agents exposes fundamental limitations in today's data management architectures. Traditional databases and data fabrics were designed for static, well-defined workloads, whereas agentic systems exhibit dynamic, context-driven, and collaborative behaviors. Agents continuously decompose tasks, shift attention across modalities, and share intermediate results with peers - producing non-deterministic, multi-modal workloads that strain conventional query optimizers and caching mechanisms. We propose an Agent-Centric Data Fabric, a unified architecture that rethinks how data systems serve, optimize, coordinate, and learn from agentic workloads. To achieve this we exploit the concepts of attention-guided data retrieval, semantic micro-caching for context-driven agent federations, predictive data prefetching and quorum-based data serving. Together, these mechanisms enable agents to access representative data faster and more efficiently, while reducing redundant queries, data movement, and inference load across systems. By framing data systems as adaptive collaborators, instead of static executors, we outline new research directions toward behaviorally responsive data infrastructures, where caching, probing, and orchestration jointly enable efficient, context-rich data exchange among dynamic, reasoning-driven agents.