This study proposes a federated learning architecture for privacy-conscious data mining in shared cloud data lakes. Using tenant-aware aggregation and decentralized model updates, it enhances prediction quality without compromising sensitive datasets, supporting regulatory compliance and intelligent enterprise decision-making.

Mohammad Idris Halim, Stefanie Ruth Malloy, Qiang Rui Zhang, Owen Douglas Thorne, Pratiksha Neelkamal Joshi, Sanjaydeep Arora

We introduce container-level scheduling heuristics designed for deploying microservices across global cloud regions. Our method optimizes environmental sustainability by factoring regional carbon intensity and energy efficiency, balancing latency and ecological footprint for next-generation multi-region cloud-native applications.

Linda Florence Eaton, Ravi Teja Kulkarni, Zihan Liu, Jean Phillippe Moreau, Fiona Bridgett McIntyre, Aadil Samarjeet Nawaz

This work proposes a hierarchical prediction model to orchestrate latency-sensitive functions in FaaS environments. The system anticipates load changes and auto-adjusts resource placement across hybrid cloud layers, ensuring seamless execution of serverless applications in real-time with improved responsiveness.

Chiang Wen Tao, Michael Leonardo Gardner, Bhavana Kalpana Reddy, Tobias Ernst Müller, Henry Joseph Lefevre

We present a VM migration approach aimed at minimizing carbon emissions during workload relocation. By considering real-time grid carbon intensity and system utilization, our strategy supports sustainable data center practices while maintaining service-level agreements in fluctuating cloud infrastructure scenarios.

Wen Xiaoqing, Javier Leonardo Montoya, Elizabeth Susan McCarthy, Rahul Brijesh Desai, Satoshi Yuuji Nakagawa

This paper outlines a cloud-edge synergy framework to assist disaster response through real-time analytics. The platform integrates AI-based image classification and edge resource pooling to improve situational awareness and reduce communication latency during emergency event coordination across cloud federations.

Ingrid Beatrix Novak, Deepak Anand Jaiswal, Yao Ming Lin, Oliver Scott Barnett, Natalia Justine Rosetti, Mahmoud Kareem Hamza

This research introduces a time-series-driven model for predictive scaling in IoT-cloud platforms. By analyzing workload patterns, the system proactively adjusts computing resources, ensuring efficient performance and cost optimization across fluctuating data influx environments supporting smart cities and industrial automation.

Mei Ling Zhao, Juan Felipe Alvarez, Haruki Shinji Mori, Angela Brooke Simmons, Nishant Devendra Patel

This paper presents a machine learning-based policy enforcement model for ensuring secure multi-tenant environments in cloud-native Kubernetes clusters. Context-aware mechanisms dynamically learn access behaviors, preventing privilege escalation and cross-tenant data exposure in containerized cloud infrastructures operating under zero-trust principles.

Daniel Trevor Hughes, Sunita Rajagopalan Menon, Zhao Cheng Wei, Francesca Pauline Grant, Emiliano Jose Paredes