AgentWare automates deploying AI agents across edge to cloud networks
AgentWare: Automating the Lifecycle of Agentic Applications across the Edge-to-Cloud Continuum
Distributed, Parallel, and Cluster ComputingMachine Learning
Summary
Running AI programs that work together on different computers from small devices to big servers is hard because these machines are very different and tricky to manage. The authors created AgentWare, a tool that automatically sets up, runs, watches, and tests these AI programs across many kinds of machines. It helps collect data about how well the AI agents work and how much energy or resources they use. They showed how AgentWare makes it easier to experiment and learn from these systems by testing a book helper AI spread across edge and cloud computers.
What this means in practice
- •For cloud infrastructure teams: Automate launching and monitoring AI agent components across mixed edge and cloud environments to improve reliability and insight.
- •For iot platform developers: Deploy distributed AI assistants on heterogeneous edge devices with less manual setup and easy performance tracking.
Authors
Michalis Kasioulis, Moysis Symeonides, George Pallis, Marios D. Dikaiakos
Abstract
Deploying LLM-enabled agentic applications across the Edge-to-Cloud continuum remains challenging due to hardware heterogeneity, deployment complexity, limited observability, and the lack of systematic evaluation methods. Existing solutions address agent development, observability, or benchmarking separately, offering limited support for the full lifecycle of distributed agentic applications. This paper presents AgentWare, an AgenticOps framework that automates the provisioning, deployment, observability, and evaluation of agentic applications across Edge-to-Cloud infrastructures. AgentWare introduces an end-to-end lifecycle pipeline that automatically prepares heterogeneous execution environments, transforms user-defined agent implementations into distributed applications, deploys agent components across the continuum, and performs unified collection of execution traces, infrastructure telemetry, and evaluation metrics. The framework further supports automated semantic evaluation through LLM-as-a-Judge workflows and generates reproducible reports covering correctness, performance, resource utilization, and energy consumption. We demonstrate the applicability of AgentWare through a distributed book assistant agent deployed across real Edge-to-Cloud infrastructure under multiple deployment and model configurations. The results show that AgentWare enables systematic experimentation and evaluation of distributed agentic applications while significantly reducing the manual effort required for deployment, instrumentation, and analysis.