This paper introduces a scalable multi-cloud solution for federated learning. It integrates privacy-preserving gradient aggregation and resource-aware distribution to support large-scale collaborative AI training across data silos while ensuring secure communication and model convergence across disparate cloud nodes.

Meera Pravin Balan, Hugo Laurence Bennett, Li Mei Qing, Ethan Maxwell Rhodes, Yuki Haruto Tanaka

This work proposes a reinforcement learning-based resource manager for SLA-aware cloud operations. The model adapts to fluctuating workload demands and optimizes virtual resource distribution, maintaining SLA compliance while reducing operational overhead and energy consumption across elastic infrastructures.

Arnav Keshav Pillai, George Edward Whitman, Huang Zhen Rong, Charlotte Irene Lawson, Jacob Thomas Greene

This research proposes a fault-tolerant framework that integrates AI monitoring for distributed cloud systems. By embedding predictive failure detection and workload migration, it minimizes system downtime and supports resilient execution of AI workloads across volatile network infrastructures.

Sanjana Nivedita Menon, Felix Noah Carter, Wen Qiang Liu, Rachel Grace Langdon, Matthew Hugo Keller

This paper presents an intelligent scaling framework using autonomous agents for fog-cloud systems. It facilitates workload distribution and auto-healing of nodes through real-time condition monitoring, enhancing fault recovery, reducing latency, and improving service uptime in hybrid deployment models.

Yong Lei Zhang, Diana Francesca Murray, Kabir Devashish Raghavan, Oliver Lucas Bradford, Leah Scarlett Jennings

This study introduces a secure identity federation model using zero-knowledge proofs for multi-tenant clouds. It eliminates credential exposure during authentication, ensuring decentralized identity trust, enhancing compliance, and mitigating attack surfaces across federated cloud ecosystems.

Yunhao Bo Wei, Natalie Faye Richardson, Rajesh Chandran Iyer, William Oscar Evans, Lucia Daniela Herrera

This research presents a serverless computing model optimized for edge-aware applications. It enhances latency-sensitive functions through stateless logic migration, reducing cold-start delays and optimizing resource reuse in geographically distributed edge-cloud systems.

Athira Sushmita Balachandran, Julian Arthur Hayes, Zhao Long Wen, Emilia Florence Woods, Timothy Charles Stewart

This study proposes an AI-powered compression framework for real-time multimedia streaming. It balances fidelity and storage efficiency by dynamically adjusting compression levels based on usage context and device constraints, supporting scalable media delivery across cloud-backed platforms.

Zhenhua Rui Zhou, Amelia Rose Hargrove, Nishant Venkatesh Krishnan, Hannah Isabelle Flynn, Logan Christopher Monroe