To enable a new generation of digital computing applications, the greatest challenge is to provide a better level of energy efficiency (intended as the performance that a system can provide within a certain power budget) without giving up a systems's flexibility. This constraint applies to digital system across all scales, starting from ultra-low power implanted devices up to datacenters for high-performance computing and for the "cloud". In this thesis, we show that architectural heterogeneity is the key to provide this efficiency and to respond to many of the challenges of tomorrow's computer architecture - and at the same time we show methodologies to introduce it with little or no loss in terms of flexibility. In particular, we show that heterogeneity can be employed to tackle the "walls" that impede further development of new computing applications: the utilization wall, i.e. the impossibility to keep all transistors on in deeply integrated chips, and the "data deluge", i.e. the amount of data to be processed that is scaling up much faster than the computing performance and efficiency. We introduce a methodology to improve heterogeneous design exploration of tightly coupled clusters; moreover we propose a fractal heterogeneity architecture that is a parallel accelerator for low-power sensor nodes, and is itself internally heterogeneous thanks to an heterogeneous coprocessor for brain-inspired computing. This platform, which is silicon-proven, can lead to more than 100x improvement in terms of energy efficiency with respect to typical computing nodes used within the same domain, enabling the application of complex algorithms, vastly more performance-hungry than the current state-of-the-art in the ULP computing domain.

Over the last years, consensus and synchronization problems have been a popular topic in the systems and control community. This interest is motivated by the fact that, in several fields of application, a certain number of agents is interacting or has to cooperate to achieve a certain task. Robotic swarms, sensor networks, power networks, biological networks are only few outstanding examples where networks of agents displays behaviors which can be modeled and studied by means of consensus and synchronisation techniques. In this thesis we consider a general class of networked nonlinear systems in different operating frameworks and design control architecture to force the systems to reach synchronization and consensus on a target behavior. In particular, we consider the case of homogeneous and heterogeneous nonlinear agents with a static communication topology and design a static high-gain-based diffusive coupling and an internal model-based regulator respectively, to solve the problem of consensus. Then, we analyze the case of dynamical links and show under which conditions, synchronization for homogeneous nonlinear systems can be achieved. Depending on the structure of the dynamic links at hand, static and dynamic regulators (based on the concept extended state observers) are proposed. Furthermore, we address the problem of disconnected topology and switching topology and derive under which conditions agents reach cluster synchronization and synchronization respectively. Last, we consider the problem of a sampled exchange of information between the agents and design a triggering rule locally at each agent such that the overall network reaches synchronization.

The Internet of Things promises to be a key-factor in the forthcoming industrial and social revolution. The Internet of Things concept rely on pervasive communications where ’things’ are ’always connected’. The focus of the thesis is on Heterogeneous Networks for Internet of Things and Machine Type Communications. Heterogeneous Networks are an enabling factor of paramount important in order to achieve the ’always connected’ paradigm. On the other hand, Machine Type Communications are deeply different from Human-to-Human communications both in terms of traffic patterns and requirements. This thesis investigate both concepts. In particular, here are studied short and long range solutions for Machine-to-machine applications. For this work a dual approach has been followed: for the short-range solutions analysis an experimental approach has been privileged; meanwhile for the long-range solutions analysis a theoretical and simulation approach has been preferred. In both case, a particular attention has been given to the feasibility of the solutions proposed, hence solutions based on products that already exist in the market have been privileged.

Performance and energy consumption in modern computing platforms is largely dominated by the memory hierarchy. The increasing computational power in the multiprocessors and accelerators, and the emergence of the data-intensive workloads (e.g. large-scale graph traversal and scientific algorithms) requiring fast transfer of large volumes of data, are two main trends which intensify this problem by putting even higher pressure on the memory hierarchy. This increasing gap between computation speed and data transfer speed is commonly referred as the “memory wall” problem. With the emergence of heterogeneous Three Dimensional (3D) Integration based on through-silicon-vias (TSV), this situation has started to recover in the past years. On one hand, it is now possible to improve memory access bandwidth and/or latency by either stacking memories directly on top of processors or through abstracted memory interfaces such as Micron’s Hybrid Memory Cube (HMC). On the other hand, near memory computation has become worthy of revisiting due to the cost-effective integration of logic and memory in 3D stacks. These two directions bring about several interesting opportunities including performance improvement, energy and cost reduction, product miniaturization, and modular design for improved time to market. In this research, we study the effectiveness of the 3D integration technology and the optimization opportunities which it can provide in the different layers of the memory hierarchy in cluster-based many-core platforms ranging from intra-cluster L1 to inter-cluster L2 scratchpad memories (SPMs), as well as the main memory. In addition, by moving a part of the computation to where data resides, in the 3D-stacked memory context, we demonstrate further energy and performance improvement opportunities.

The objective of this thesis is twofold: on one hand, the design of nonliner observers, on the other, the design of internal-model regulators to solve the robust output regulation problem. In the observer theory a key role is played by the so called high-gain observers. The purpose of the first part of the thesis is to propose novel techniques which allow to overcome or at least to mitigate some of the main drawbacks characterizing this class of observers. Firstly, we study the possibility of writing an observer for multi-input multi-output observable systems in the original coordinates. Then, we propose a novel class of high-gain observers, denoted as ``low-power'', which allows to overcome numerical problems, to avoid the peaking phenomenon and to improve the sensitivity properties to high-frequency measurement noise. The second part of the thesis addresses the output regulation problem, solved for linear systems during the 70's by Francis and Wonham who coined the celebrated ``internal model principle''. Constructive solutions have also been proposed in the nonlinear framework but under restrictive assumptions that reduce the class of systems to which this methodology can be applied. In this thesis we focus on the output regulation problem in presence of periodic disturbances and we propose a novel approach which allows to consider a broader class of nonlinear systems. With the proposed design the stabilization problem and the regulation problem are substantially decoupled and output regulation is achieved in presence of uncertainties or disturbances, as long as the trajectories of the resulting closed-loop system are bounded.

In this thesis, the author presents a query language for an RDF (Resource Description Framework) database and discusses its applications in the context of the HELM project (the Hypertextual Electronic Library of Mathematics). This language aims at meeting the main requirements coming from the RDF community. in particular it includes: a human readable textual syntax and a machine-processable XML (Extensible Markup Language) syntax both for queries and for query results, a rigorously exposed formal semantics, a graph-oriented RDF data access model capable of exploring an entire RDF graph (including both RDF Models and RDF Schemata), a full set of Boolean operators to compose the query constraints, fully customizable and highly structured query results having a 4-dimensional geometry, some constructions taken from ordinary programming languages that simplify the formulation of complex queries. The HELM project aims at integrating the modern tools for the automation of formal reasoning with the most recent electronic publishing technologies, in order create and maintain a hypertextual, distributed virtual library of formal mathematical knowledge. In the spirit of the Semantic Web, the documents of this library include RDF metadata describing their structure and content in a machine-understandable form. Using the author's query engine, HELM exploits this information to implement some functionalities allowing the interactive and automatic retrieval of documents on the basis of content-aware requests that take into account the mathematical nature of these documents.

Biological data are inherently interconnected: protein sequences are connected to their annotations, the annotations are structured into ontologies, and so on. While protein-protein interactions are already represented by graphs, in this work I am presenting how a graph structure can be used to enrich the annotation of protein sequences thanks to algorithms that analyze the graph topology. We also describe a novel solution to restrict the data generation needed for building such a graph, thanks to constraints on the data and dynamic programming. The proposed algorithm ideally improves the generation time by a factor of 5. The graph representation is then exploited to build a comprehensive database, thanks to the rising technology of graph databases. While graph databases are widely used for other kind of data, from Twitter tweets to recommendation systems, their application to bioinformatics is new. A graph database is proposed, with a structure that can be easily expanded and queried.

A Smart City is a high-performance urban context, where citizens live independently and are more aware of the surrounding opportunities, thanks to forward-looking development of economy politics, governance, mobility and environment. ICT infrastructures play a key-role in this new research field being also a mean for society to allow new ideas to prosper and new, more efficient approaches to be developed. The aim of this work is to research and develop novel solutions, here called smart services, in order to solve several upcoming problems and known issues in urban areas and more in general in the modern society context. A specific focus is posed on smart governance and on privacy issues which have been arisen in the cellular age.

Recent research has shown that the performance of a single, arbitrarily efficient algorithm can be significantly outperformed by using a portfolio of —possibly on-average slower— algorithms. Within the Constraint Programming (CP) context, a portfolio solver can be seen as a particular constraint solver that exploits the synergy between the constituent solvers of its portfolio for predicting which is (or which are) the best solver(s) to run for solving a new, unseen instance. In this thesis we examine the benefits of portfolio solvers in CP. Despite portfolio approaches have been extensively studied for Boolean Satisfiability (SAT) problems, in the more general CP field these techniques have been only marginally studied and used. We conducted this work through the investigation, the analysis and the construction of several portfolio approaches for solving both satisfaction and optimization problems. We focused in particular on sequential approaches, i.e., single-threaded portfolio solvers always running on the same core. We started from a first empirical evaluation on portfolio approaches for solving Constraint Satisfaction Problems (CSPs), and then we improved on it by introducing new data, solvers, features, algorithms, and tools. Afterwards, we addressed the more general Constraint Optimization Problems (COPs) by implementing and testing a number of models for dealing with COP portfolio solvers. Finally, we have come full circle by developing sunny-cp: a sequential CP portfolio solver that turned out to be competitive also in the MiniZinc Challenge, the reference competition for CP solvers.

Until few years ago, 3D modelling was a topic confined into a professional environment. Nowadays technological innovations, the 3D printer among all, have attracted novice users to this application field. This sudden breakthrough was not supported by adequate software solutions. The 3D editing tools currently available do not assist the non-expert user during the various stages of generation, interaction and manipulation of 3D virtual models. This is mainly due to the current paradigm that is largely supported by two-dimensional input/output devices and strongly affected by obvious geometrical constraints. We have identified three main phases that characterize the creation and management of 3D virtual models. We investigated these directions evaluating and simplifying the classic editing techniques in order to propose more natural and intuitive tools in a pure 3D modelling environment. In particular, we focused on freehand sketch-based modelling to create 3D virtual models, interaction and navigation in a 3D modelling environment and advanced editing tools for free-form deformation and objects composition. To pursuing these goals we wondered how new gesture-based interaction technologies can be successfully employed in a 3D modelling environments, how we could improve the depth perception and the interaction in 3D environments and which operations could be developed to simplify the classical virtual models editing paradigm. Our main aims were to propose a set of solutions with which a common user can realize an idea in a 3D virtual model, drawing in the air just as he would on paper. Moreover, we tried to use gestures and mid-air movements to explore and interact in 3D virtual environment, and we studied simple and effective 3D form transformations. The work was carried out adopting the discrete representation of the models, thanks to its intuitiveness, but especially because it is full of open challenges.