Processing and Learning Orchestrator

Fondazione LINKS – Leading Innovation & Knowledge for Society

This component manages the lifecycle of AI models and monitors their performance across different deployment environments.

The Processing and Learning Orchestrator component is designed for deploying AI models in cloud and fog environments, offering REST APIs for model lifecycle management. It features Deployment Agents for CRUD operations on models and an internal worker for task execution. Additionally, it includes a monitoring stack for tracking hardware resources, model resource consumption, and learning performance, accessible via APIs and Grafana dashboards.