Renewable energy forecasting—particularly for distributed solar power—is vital for ensuring grid reliability and efficient resource management. However, traditional centralized forecasting approaches often come with drawbacks like high communication overhead, privacy issues, and limited scalability. Federated Learning (FL), consists of a decentralized machine learning framework that allows edge devices to collaboratively train models without disclosing raw data. Yet, FL struggles when devices have limited or sparse data, leading to cold-start problems and weaker model performance. To mitigate these challenges, this work introduces a method that incorporates GAN-based data augmentation to enrich local datasets with synthetic time-series samples during the initialization phase of the federated training process.

High-performance computing (HPC) and cloud platforms provide complementary capabilities for scientific workflows, yet their integration remains challenging due to differences in resource management, data movement, security requirements, and hardware heterogeneity. These limitations hinder the realization of seamless hybrid computing within federated European environments. This paper introduces NOUS, a modular framework designed to enable HPC–cloud interoperability. The architecture is structured into four layers: user interfaces (e.g., Jupyter, Open OnDemand), workflow orchestration, resource management via the SLURM REST API, and container portability through Apptainer. The framework incorporates federated authentication, locality-aware data management, and standardized interfaces to support secure and compliant hybrid execution, including cloud bursting.

Modern research infrastructures increasingly span heterogeneous environments, including cloud computing, high-performance computing (HPC), edge computing nodes, and most recently quantum computers (QC). This diversity challenges traditional perimeter-based security models and demands more agile, robust cybersecurity solutions. In response, we present a minimal yet comprehensive cybersecurity architecture built around Zero Trust principles and darknet-inspired principles. Our solution adopts a deployment-agnostic approach, abstracting away environmental differences through encrypted overlay networks and identity-driven authentication and authorization mechanisms.

Federated Learning (FL) has increasingly been promoted as a practical way to train models over distributed edge data while avoiding raw-data centralization, yet most published evaluations have remained confined to a single task and therefore have provided limited evidence about how FL behaves across substantially different workloads. In this paper, a multi-domain empirical study of FL is presented in two representative edge settings: vision-based object detection for connected vehicles and time-series energy forecasting for smart-grid assets.

This report presents the approach adopted within WP7.1 for collecting and structuring data and observations derived from NOUS validation sites and business simulation activities. The work is aligned with the objective of establishing a quantification framework for the data economy, supporting the assessment of data value across different application domains. Within the NOUS project, validation activities generate heterogeneous evidence, including technical performance data, operational feedback, and user-related insights. In parallel, business simulations provide an additional layer of analysis, enabling the exploration of economic sustainability and scalability under controlled conditions. The challenge addressed in this task is therefore not only to collect such information, but to transform it into structured knowledge supporting decision-making and business model development.

This white paper presents an overview of the current state of the art in high-performance computing (HPC) and its convergence with cloud technologies, with a strategic focus on innovation management and exploitation. It outlines recent advances in cloud-based HPC services, hybrid architectures, and federated systems that integrate edge, cloud, and HPC resources. Emerging paradigms such as federated learning, AIdriven optimization, and sustainable computing are analyzed for their transformative potential. Special emphasis is placed on the NOUS project, which exemplifies a holistic approach to federated HPC-cloud services, combining technical innovation with robust exploitation strategies. NOUS goes one step beyond and explores the integration of quantum computing in handling data existing in the cloud. As of late 2025, the synergy between Cloud Computing and Quantum Computing (QC) has matured into a functional “Quantum-as-a-Service” (QaaS) model. While physical quantum hardware remains too fragile for on-premise deployment, cloud providers have democratized access to the “Quantum Stack.” This report highlights this progress, the issues and real future applications. NOUS addresses European priorities for digital sovereignty and data interoperability, supporting scalable, privacy-preserving, and AI-enabled HPC workflows. The paper concludes with a roadmap that positions NOUS as a reference architecture and innovation catalyst for Europe’s distributed computing ecosystem.