This study introduces a swarm-optimized allocation algorithm for managing multitenant cloud workloads. It improves task segregation, reduces resource contention, and adapts to fluctuating demands. Results demonstrate efficiency gains and load balancing across hybrid cloud platforms using simulated multi-objective performance baselines.

Krishna Dinesh Raghunathan, Javier Emmanuel Lopez, Alina Beatrice Cristea, Chen Zhao Ming, Timothy Blake Armstrong

A novel machine learning-driven method for energy-efficient migration of containers across edge, fog, and cloud layers is presented. It reduces downtime, optimizes resource usage, and adapts to shifting workload conditions, enabling intelligent distribution of services in time-critical environments.

Fatima Zahra Nouri, Yuki Hiroto Nakamura, Siddharth Rajeev Kapoor, Emily Rose Beckett, Zhao Liang Sheng

This paper introduces an intent-based framework to manage microservices across heterogeneous cloud platforms. It simplifies orchestration, enhances scalability, and ensures compliance through automated policy enforcement, demonstrating effective load balancing and resilience under varying service conditions.

Rakesh Vimal Menon, Harriet Olivia Foster, Junichi Tanaka Shiro, Clara Devereux Smith, Matteo Riccardo Bianciardi

We propose a federated learning framework designed with built-in security for cloud networks. It ensures data confidentiality while enabling collaborative model training, offering a privacy-preserving solution for distributed AI systems operating across jurisdictionally diverse cloud environments.

Nia Sophia Allen, Wei Chao Lin, Arun Kumar Natarajan, Hailey Joana Marques, Kristof Emil Sorensen

A scheduling algorithm optimized for smart city cloud systems is proposed. It integrates real-time sensor data with cloud workloads to minimize latency, improve throughput, and support multimodal data analysis in time-sensitive urban monitoring applications.

Mikhail Olegovich Petrov, Shravan Deepak Pillai, George Alvin Blackwell, Yuting Lianhua Zhang, Pascal Rene Lafleur

The model forecasts compute demand in dynamic cloud environments using deep learning ensembles. It supports proactive scaling, reduces operational costs, and maintains SLA compliance by predicting workload trends across diverse cloud-native service configurations.

Jacqueline Maria Thompson, Hao Rui Li, Karthik Venkatraman Iyer, Jean Louis Picard, Eva Helene Schmitt

This study develops an edge-optimized AI pipeline for analyzing video streams in cloud surveillance systems. It reduces bandwidth usage and enhances detection speed, providing a scalable solution for real-time anomaly recognition in high-traffic environments.

Junaid Bashir Qureshi, Xiao Ming Zhu, Kimberly Ann Doyle, Andre Luis Henriquez, Naomi Ruth Goldberg