Blockchain Digital Twin approach to modeling HPC Data Center
Please login to view abstract download link
Data Center (DC) 's energy consumption has skyrocketed due to the surge in digital technology, services, and associated applications. By 2030, they are expected to consume 13% of the global energy demand, potentially up to 21%, and release 8% of the associated global carbon emissions. This can lead to significant operational costs, power security impacts, and environmental threats. Prior heuristics, statistics, and engineering approaches are ineffective in improving energy efficiency because of the sheer number of configurations, non-linear parameter interactions, and vast amounts of operating data. Hence, optimizing a DC, particularly the class of High-Performance Computing (HPC) clusters, is a significant concern. Integrating the IoT, sensors, and intelligent devices has significantly contributed to generating vast operational management data from various aspects of the DC industry. Indeed, effectively modeling and processing this data could improve energy efficiency, ensure reliability, reduce operating costs, and sustainably manage DC. This study proposes a holistic approach based on Blockchain Digital Twin, augmented by Artificial Intelligence (AI) processed data to represent the DC physical complex system in virtual and tokenized models. The objective is to obtain near-real-time prediction, optimization, monitoring, controlling and improved decision-making. In detail, the methodology, the architecture and steps taken to develop a scalable analytical dashboard in Blockchain are shown. The pioneering Blockchain DC Digital Twin approach is experimented with in a real infrastructural setting; ENEA HPC-DC learns from actual operational data and allows the management, monitoring, and control in real-time visualization of the whole DC, thus allowing coverage of the DC status and the optimization of its functionalities. The authors also aim to show the achieved benefits of the Blockchain Digital Twin DC challenge for various stakeholders.
