This paper introduces an adaptive scaling framework tailored for latency-sensitive services in multi-tier cloud architectures. It leverages traffic-aware prediction models to dynamically allocate resources, ensuring consistent performance under fluctuating demand while minimizing overhead across microservices and backend layers.

Julian Scott Holloway, Amelia Dawn Kensington, Noah Gabriel Fletcher, Penelope Sarah Aldridge, Theodore James Pemberton

This paper explores a blockchain-enabled access management protocol for decentralized cloud data lakes. It secures sensitive data through immutable smart contracts and provides audit trails for regulatory compliance, ensuring secure, transparent data sharing across organizational boundaries in collaborative environments.

Charlotte Emily Harrington, Benjamin Oliver Prescott, Abigail Elise Stone, Finley George Radcliffe, Henry Thomas Blake

This research presents a redundancy-aware disaster recovery model using virtual replication for critical cloud applications. It ensures business continuity by enabling real-time failover, geographically isolated backup systems, and intelligent prioritization of resource provisioning during catastrophic outages.

Joseph Maxwell Kendall, Clara Isabelle Morton, Samuel Edward Baines, Lydia Florence Carrington, Hugo William Dorsey

This paper proposes a genetic algorithm framework to configure cloud services with trade-offs between performance and cost. The model generates Pareto-optimal deployment plans, balancing computing efficiency, latency, and budget constraints in dynamic and heterogeneous cloud application environments.

Frederick Anthony Weston, Georgia Mae Pennington, Patrick John Collier, Eleanor Beatrice Langford, Sebastian Isaac Menzies

This research demonstrates a serverless workflow engine for orchestrating high-throughput scientific computations. It supports diverse functions across cloud platforms using an event-driven scheduler, optimizing execution order, cold start mitigation, and minimizing billing overhead through fine-grained resource allocation and parallel invocation strategies.

Leonard Christopher Pryce, Florence Rebecca Wilton, Nathaniel Jack Hollingsworth, Alice Olivia Tremayne, Oscar Philip Rowe

This paper proposes a deep learning-based load balancing system for federated cloud environments. It models inter-node latency using graph neural networks, enabling intelligent traffic routing and minimizing bottlenecks while optimizing user experience in multi-cloud distributed applications.

Gregory Allen Whitman, Harriet Louise Mansfield, Vincent Charles Goodwin, Elise Victoria Thornton, Dominic Lewis Farrington

This study introduces a predictive migration strategy for containerized applications across edge-cloud layers. Using real-time resource heatmaps and machine learning models, the system anticipates congestion and migrates workloads proactively to maintain service quality and balance compute demand.

Matilda Jane Hargreaves, Hugo Leonard Fenton, Isabelle Claire Huxley, Zachary Felix Browning, Ava Rose Dunnington