This study presents a federated learning framework using knowledge distillation for privacy-preserving training across heterogeneous cloud-edge environments. The system supports dynamic participation, ensuring robust model aggregation without sharing raw data, while maintaining convergence performance on non-iid and bandwidth-constrained deployments.

Meera Ananth Srinivas, Jonathan Felix Carter, Yao Ming Cheng, Sophia Helen Redding, Luca Manuel Romero

This paper proposes a latency-aware VM placement algorithm that enhances energy efficiency in green data centers. The approach balances workload distribution and thermal footprint while minimizing communication delays between interdependent services in large-scale cloud infrastructures.

Grace Matilda Blake, Ravi Tejas Gopalakrishnan, Ethan Joseph Palmer, Ji Wen Xiu, Laura Isabelle McConnell

This research introduces a behavior tree-based self-adaptive intrusion response system for multi-tenant cloud platforms. It models cloud service behaviors and enacts dynamic, context-driven security responses with reduced false positives and recovery overhead in distributed threat scenarios.

Xin Liang Bo, Hannah Claire Davies, Adarsh Shankar Rajan, Zachary Finn Harper, Mila Adriana Vasquez

This paper presents an AI-based workflow partitioning system that decomposes scientific data pipelines for execution across distributed microservices. It improves latency, fault tolerance, and computational throughput by dynamically rebalancing workloads according to task complexity and cloud resource metrics.

Alejandro Ramon Silva, Li Jia Nan, Priyanka Devi Nair, George Nicholas Stokes, Clara Emmeline Foster

This study proposes a blockchain-backed cloud storage solution for medical image archives. It ensures tamper-proof logging and access auditing through encrypted transaction records and timestamped validation, supporting regulatory compliance in cross-jurisdictional medical cloud applications.

Charlotte Mae Dawson, Deepak Ramesh Varma, Xiang Yuan Zhe, Benjamin Oscar Nolan, Gianna Lucia Ortega

This research presents a multimodal AI framework that aggregates distributed monitoring streams to assess cloud application health. It correlates logs, metrics, and traces with contextual triggers to proactively identify bottlenecks, anomalies, and performance degradations across complex deployments.

Arjun Vinayak Rao, Matilda Elise Thornton, Zhao Wei Long, Henry Nathaniel Lloyd, Sophia Adriana Hayes