Self-learning based Decentralized Resource Management of Cloud Continuum

This poster focuses on self-learning-based decentralized resource management in the Cloud Continuum, emphasizing the challenges and complexities of managing resources across distributed computing systems. It addresses the need for efficient, scalable, and secure management methods due to the dynamic, heterogeneous, and geographically distributed nature of cloud and edge environments.

Emerging applications like IoT, smart cities, digital twins, and VR streaming require responsive and adaptable resource management solutions. The work explores multi-objective optimization approaches using machine learning and reinforcement learning, emphasizing decentralized control to optimize performance and ensure privacy and security within this distributed ecosystem.