Monitoring and Prediction of Thermal Emergencies in High Performance Computing Systems

Modern scientific discoveries are driven by an unsatisfiable demand for computational resources. High-Performance Computing (HPC) systems are an aggregation of computing power to deliver considerably higher performance than one typical desktop computer can provide, to solve large problems in science, engineering, or business. An HPC room in the datacenter is a complex controlled environment that hosts thousands of computing nodes that consume electrical power in the range of megawatts, which gets completely transformed into heat. Although a datacenter contains sophisticated cooling systems, our studies indicate quantitative evidence of thermal bottlenecks in real-life production workload, showing the presence of significant spatial and temporal thermal and power heterogeneity. Therefore minor thermal issues/anomalies can potentially start a chain of events that leads to an unbalance between the amount of heat generated by the computing nodes and the heat removed by the cooling system originating thermal hazards. Although thermal anomalies are rare events, anomaly detection/prediction in time is vital to avoid IT and facility equipment damage and outage of the datacenter, with severe societal and business losses. For this reason, automated approaches to detect thermal anomalies in datacenters have considerable potential. This thesis analyzed and characterized the power and thermal characteristics of a Tier0 datacenter (CINECA) during production and under abnormal thermal conditions. Then, a Deep Learning (DL)-powered thermal hazard prediction framework is proposed. The proposed models are validated against real thermal hazard events reported for the studied HPC cluster while in production. This thesis is the first empirical study of thermal anomaly detection and prediction techniques of a real large-scale HPC system to the best of my knowledge. For this thesis, I used a large-scale dataset, monitoring data of tens of thousands of sensors for around 24 months with a data collection rate of around 20 seconds.

PCM-based battery thermal management systems, energy saving materials and LCA strategy for automotive

This PhD work arises from the necessity to give a contribution to the energy saving field, regarding automotive applications. The aim was to produce a multidisciplinary work to show how much important is to consider different aspects of an electric car realization: from innovative materials to cutting-edge battery thermal management systems (BTMSs), also dealing with the life cycle assessment (LCA) of the battery packs (BPs). Regarding the materials, it has been chosen to focus on carbon fiber composites as their use allows realizing light products with great mechanical properties. Processes and methods to produce carbon fiber goods have been analysed with a special attention on the university solar car Emilia 4. The work proceeds dealing with the common BTMSs on the market (air-cooled, cooling plates, heat pipes) and then it deepens some of the most innovative systems such as the PCM-based BTMSs after a previous experimental campaign to characterize the PCMs. After that, a complex experimental campaign regarding the PCM-based BTMSs has been carried on, considering both uninsulated and insulated systems. About the first category the tested systems have been pure PCM-based and copper-foam-loaded-PCM-based BTMSs; the insulated tested systems have been pure PCM-based and copper-foam-loaded-PCM-based BTMSs and both of these systems equipped with a liquid cooling circuit. The choice of lighter building materials and the optimization of the BTMS are strategies which helps in reducing the energy consumption, considering both the energy required by the car to move and the BP state of health (SOH). Focusing on this last factor, a clear explanation regarding the importance of taking care about the SOH is given by the analysis of a BP production energy consumption. This is why a final dissertation about the life cycle assessment (LCA) of a BP unit has been presented in this thesis.

Multidisciplinary approach to horse welfare in different management systems

From its domestication until nowadays, the horse has assumed multiple roles in human society. Over time, this condition and the lack of specific regulation have led to the development of different kinds of management systems for this species. This Ph.D. research project aims to investigate horses' welfare in different management practices and housing systems, considering a multidisciplinary approach, taking into account biological function, naturalness, and affective dimension. The results are presented in five articles that evidence risk factors that can mine horse welfare, and examine tools and parameters that can be employed for its assessment. Our research shows the importance of considering the evolutionary history and the species-specific and behavioural needs of horses in their management and housing. Sociality, free movement, diet composition and foraging routine, and the workload that these animals undergo are important factors that should be taken into account. Furthermore, this research has evidenced the importance of employing different parameters (e.g., behaviour, endocrinological parameters, and immune activity) in welfare assessment and proposes the use of horsehair DHEA (dehydroepiandrosterone) as a possible useful additional non-invasive measure for the investigation of long-term stress conditions. Finally, our results underline the importance of considering the affective dimension in welfare research. Recently, Judgement Bias Tests (JBT), which are based on the influence of affective states on the decision-making process, have been widely employed in animal welfare research. However, our studies show that the use of spatial JBT in horses can have some limitations. Still today several management systems do not fulfill species-specific needs of horses, thus the implementation of specific regulations could ameliorate horse welfare. A multidisciplinary approach to welfare assessment is fundamental, but it should be always remembered the individual and its own characteristics, which can influence not only physiological, immunological, and behavioural responses but also emotional and cognitive dimensions.

Dynamic power management: from portable devices to high performance computing

Electronic applications are nowadays converging under the umbrella of the cloud computing vision. The future ecosystem of information and communication technology is going to integrate clouds of portable clients and embedded devices exchanging information, through the internet layer, with processing clusters of servers, data-centers and high performance computing systems. Even thus the whole society is waiting to embrace this revolution, there is a backside of the story. Portable devices require battery to work far from the power plugs and their storage capacity does not scale as the increasing power requirement does. At the other end processing clusters, such as data-centers and server farms, are build upon the integration of thousands multiprocessors. For each of them during the last decade the technology scaling has produced a dramatic increase in power density with significant spatial and temporal variability. This leads to power and temperature hot-spots, which may cause non-uniform ageing and accelerated chip failure. Nonetheless all the heat removed from the silicon translates in high cooling costs. Moreover trend in ICT carbon footprint shows that run-time power consumption of the all spectrum of devices accounts for a significant slice of entire world carbon emissions. This thesis work embrace the full ICT ecosystem and dynamic power consumption concerns by describing a set of new and promising system levels resource management techniques to reduce the power consumption and related issues for two corner cases: Mobile Devices and High Performance Computing.

Power flow forecasting and smart charging for electric vehicle charging systems with photovoltaic integration

An essential role in the global energy transition is attributed to Electric Vehicles (EVs) the energy for EV traction can be generated by renewable energy sources (RES), also at a local level through distributed power plants, such as photovoltaic (PV) systems. However, EV integration with electrical systems might not be straightforward. The intermittent RES, combined with the high and uncontrolled aggregate EV charging, require an evolution toward new planning and paradigms of energy systems. In this context, this work aims to provide a practical solution for EV charging integration in electrical systems with RES. A method for predicting the power required by an EV fleet at the charging hub (CH) is developed in this thesis. The proposed forecasting method considers the main parameters on which charging demand depends. The results of the EV charging forecasting method are deeply analyzed under different scenarios. To reduce the EV load intermittency, methods for managing the charging power of EVs are proposed. The main target was to provide Charging Management Systems (CMS) that modulate EV charging to optimize specific performance indicators such as system self-consumption, peak load reduction, and PV exploitation. Controlling the EV charging power to achieve specific optimization goals is also known as Smart Charging (SC). The proposed techniques are applied to real-world scenarios demonstrating performance improvements in using SC strategies. A viable alternative to maximize integration with intermittent RES generation is the integration of energy storage. Battery Energy Storage Systems (BESS) may be a buffer between peak load and RES production. A sizing algorithm for PV+BESS integration in EV charging hubs is provided. The sizing optimization aims to optimize the system's energy and economic performance. The results provide an overview of the optimal size that the PV+BESS plant should have to improve whole system performance in different scenarios.

Development and testing of methodologies for the control and diagnosis of innovative combustion systems

In pursuit of aligning with the European Union's ambitious target of achieving a carbon-neutral economy by 2050, researchers, vehicle manufacturers, and original equipment manufacturers have been at the forefront of exploring cutting-edge technologies for internal combustion engines. The introduction of these technologies has significantly increased the effort required to calibrate the models implemented in the engine control units. Consequently the development of tools that reduce costs and the time required during the experimental phases, has become imperative. Additionally, to comply with ever-stricter limits on 〖"CO" 〗_"2" emissions, it is crucial to develop advanced control systems that enhance traditional engine management systems in order to reduce fuel consumption. Furthermore, the introduction of new homologation cycles, such as the real driving emissions cycle, compels manufacturers to bridge the gap between engine operation in laboratory tests and real-world conditions. Within this context, this thesis showcases the performance and cost benefits achievable through the implementation of an auto-adaptive closed-loop control system, leveraging in-cylinder pressure sensors in a heavy-duty diesel engine designed for mining applications. Additionally, the thesis explores the promising prospect of real-time self-adaptive machine learning models, particularly neural networks, to develop an automatic system, using in-cylinder pressure sensors for the precise calibration of the target combustion phase and optimal spark advance in a spark-ignition engines. To facilitate the application of these combustion process feedback-based algorithms in production applications, the thesis discusses the results obtained from the development of a cost-effective sensor for indirect cylinder pressure measurement. Finally, to ensure the quality control of the proposed affordable sensor, the thesis provides a comprehensive account of the design and validation process for a piezoelectric washer test system.

Applicazione di strumenti e metodi per la gestione ambientale di un territorio vasto - Sviluppo di un sistema di gestione ambientale in un ente pubblico secondo lo schema europeo EMAS e analisi di un sistema energetico territoriale

Environmental Management includes many components, among which we can include Environmental Management Systems (EMS), Environmental Reporting and Analysis, Environmental Information Systems and Environmental Communication. In this work two applications are presented: the developement and implementation of an Environmental Management System in local administrations, according to the European scheme "EMAS", and the analysis of a territorial energy system through scenario building and environmental sustainability assessment. Both applications are linked by the same objective, which is the quest for more scientifically sound elements; in fact, both EMS and energy planning are oftec carachterized by localism and poor comparability. Emergy synthesis, proposed by ecologist H.T. Odum and described in his book "Environmental Accounting: Emergy and Environmental Decision Making" (1996) has been chosen and applied as an environmental evaluation tool, in order complete the analysis with an assessment of the "global value" of goods and processes. In particular, eMergy syntesis has been applied in order to improve the evaluation of the significance of environmental aspects in an EMS, and in order to evaluate the environmental performance of three scenarios of future evolution of the energy system. Regarding EMS, in this work an application of an EMS together with the CLEAR methodology for environmental accounting is discussed, in order to improve the identification of the environmental aspects; data regarding environmental aspects and significant ones for 4 local authorities are also presented, together with a preliminary proposal for the integration of the assessment of the significance of environmental aspects with eMergy synthesis. Regarding the analysis of an energy system, in this work the carachterization of the current situation is presented together with the overall energy balance and the evaluation of the emissions of greenhouse gases; moreover, three scenarios of future evolution are described and discussed. The scenarios have been realized with the support of the LEAP software ("Long Term Energy Alternatives Planning System" by SEI - "Stockholm Environment Institute"). Finally, the eMergy synthesis of the current situation and of the three scenarios is shown.

QoS-aware architectures, technologies, and middleware for the cloud continuum

The recent trend of moving Cloud Computing capabilities to the Edge of the network is reshaping how applications and their middleware supports are designed, deployed, and operated. This new model envisions a continuum of virtual resources between the traditional cloud and the network edge, which is potentially more suitable to meet the heterogeneous Quality of Service (QoS) requirements of diverse application domains and next-generation applications. Several classes of advanced Internet of Things (IoT) applications, e.g., in the industrial manufacturing domain, are expected to serve a wide range of applications with heterogeneous QoS requirements and call for QoS management systems to guarantee/control performance indicators, even in the presence of real-world factors such as limited bandwidth and concurrent virtual resource utilization. The present dissertation proposes a comprehensive QoS-aware architecture that addresses the challenges of integrating cloud infrastructure with edge nodes in IoT applications. The architecture provides end-to-end QoS support by incorporating several components for managing physical and virtual resources. The proposed architecture features: i) a multilevel middleware for resolving the convergence between Operational Technology (OT) and Information Technology (IT), ii) an end-to-end QoS management approach compliant with the Time-Sensitive Networking (TSN) standard, iii) new approaches for virtualized network environments, such as running TSN-based applications under Ultra-low Latency (ULL) constraints in virtual and 5G environments, and iv) an accelerated and deterministic container overlay network architecture. Additionally, the QoS-aware architecture includes two novel middlewares: i) a middleware that transparently integrates multiple acceleration technologies in heterogeneous Edge contexts and ii) a QoS-aware middleware for Serverless platforms that leverages coordination of various QoS mechanisms and virtualized Function-as-a-Service (FaaS) invocation stack to manage end-to-end QoS metrics. Finally, all architecture components were tested and evaluated by leveraging realistic testbeds, demonstrating the efficacy of the proposed solutions.