This paper explores Bayesian optimization for intelligent resource allocation in cloud platforms. It integrates predictive performance modelling to anticipate workload demands, improving utilization efficiency, reducing SLA violations, and maintaining consistent application performance in variable demand environments.

Claudia Renee Whittaker, Simon Elliot Bradley, Fiona Madeleine Granger, Gerald Anthony Mitchell, Olivia Jeanette Dawson

The study presents a container orchestration strategy for hybrid clouds using energy-aware policies. It optimizes microservice deployment in Kubernetes clusters, balancing workload distribution and energy efficiency by analyzing thermal footprints and compute intensity across private and public nodes.

Julian Christopher Merrick, Edith Valerie Donnelly, Henry Malcolm Croft, Louisa Bridget Hartley, Nathaniel Grant Fowler

This paper introduces a privacy-focused machine learning architecture for federated cloud systems. It combines differential privacy with secure multiparty computation, enabling collaborative training on decentralized data while safeguarding sensitive user information and complying with global privacy regulations.

Frederick James Monaghan, Eleanor Patricia Rowe, Calvin Douglas Finch, Rebecca Lynn Cartwright, Stephen Douglas Redmond

The paper introduces a graph-based optimization model for orchestrating serverless functions. It identifies performance bottlenecks and latency constraints through DAG analysis, enabling dynamic scheduling and load balancing for cost-effective and efficient function execution across multicloud environments.

Gareth Lionel Prescott, Marianne Judith Langley, Oscar Trevor Jenkins, Penelope Abigail Norris, Edwin Howard Sinclair

This research builds a cloud-native analytics platform tailored for smart cities. It processes high-frequency urban mobility streams with real-time analytics and edge-enhanced event correlation, offering insights into traffic trends, commuter behavior, and urban planning needs through scalable distributed infrastructures.

Isabelle Florence Murray, Douglas Charles Whitaker, Lydia Harriet Campbell, Vincent Arthur Holland, Charlotte Jane Foster

This paper proposes a dynamic load balancing algorithm that improves energy efficiency in data centers. It reallocates virtual machines based on predictive analysis of workload spikes, thereby reducing energy wastage and optimizing computational throughput in large-scale cloud infrastructures.

Vivian Theresa McCormick, Jonathan Lewis Goodwin, Abigail Rose Stanley, Bernard Keith Holloway, Sylvia Diane Mathews

The paper introduces a predictive maintenance model leveraging edge-to-cloud integration. It enables fault prediction in industrial IoT systems by analyzing sensor data in real time, improving operational uptime and reducing downtime through early anomaly detection and adaptive model updates.

Harold Vincent Sanders, Bridget Elaine Wallace, Miles Franklin Kerr, Evelyn Frances Dorsey, Kenneth Nathaniel Brooks