Use of shared memory in the context of embedded multi-core processors: exploration of the technology and its limits

Modern embedded systems embrace many-core shared-memory designs. Due to constrained power and area budgets, most of them feature software-managed scratchpad memories instead of data caches to increase the data locality. It is therefore programmers’ responsibility to explicitly manage the memory transfers, and this make programming these platform cumbersome. Moreover, complex modern applications must be adequately parallelized before they can the parallel potential of the platform into actual performance. To support this, programming languages were proposed, which work at a high level of abstraction, and rely on a runtime whose cost hinders performance, especially in embedded systems, where resources and power budget are constrained. This dissertation explores the applicability of the shared-memory paradigm on modern many-core systems, focusing on the ease-of-programming. It focuses on OpenMP, the de-facto standard for shared memory programming. In a first part, the cost of algorithms for synchronization and data partitioning are analyzed, and they are adapted to modern embedded many-cores. Then, the original design of an OpenMP runtime library is presented, which supports complex forms of parallelism such as multi-level and irregular parallelism. In the second part of the thesis, the focus is on heterogeneous systems, where hardware accelerators are coupled to (many-)cores to implement key functional kernels with orders-of-magnitude of speedup and energy efficiency compared to the “pure software” version. However, three main issues rise, namely i) platform design complexity, ii) architectural scalability and iii) programmability. To tackle them, a template for a generic hardware processing unit (HWPU) is proposed, which share the memory banks with cores, and the template for a scalable architecture is shown, which integrates them through the shared-memory system. Then, a full software stack and toolchain are developed to support platform design and to let programmers exploiting the accelerators of the platform. The OpenMP frontend is extended to interact with it.

Extending Implicit Computational Complexity and Abstract Machines to Languages with Control

The Curry-Howard isomorphism is the idea that proofs in natural deduction can be put in correspondence with lambda terms in such a way that this correspondence is preserved by normalization. The concept can be extended from Intuitionistic Logic to other systems, such as Linear Logic. One of the nice conseguences of this isomorphism is that we can reason about functional programs with formal tools which are typical of proof systems: such analysis can also include quantitative qualities of programs, such as the number of steps it takes to terminate. Another is the possiblity to describe the execution of these programs in terms of abstract machines. In 1990 Griffin proved that the correspondence can be extended to Classical Logic and control operators. That is, Classical Logic adds the possiblity to manipulate continuations. In this thesis we see how the things we described above work in this larger context.

Il linguaggio figurato e la cabina di simultanea: Il progetto IMITES(Interpretación de la Metáfora entre ITaliano y ESpañol)

Il progetto di dottorato IMITES (Interpretación de la Metáfora entre ITaliano y ESpañol) si pone come obiettivo quello di analizzare l’interpretazione simultanea del linguaggio figurato nelle combinazioni italiano-spagnolo e spagnolo-italiano. Prevede l’analisi di una serie di dati estratti da discorsi pronunciati in italiano e spagnolo in occasione di conferenze tenutesi presso la Commissione europea, e le loro versioni interpretate in spagnolo e italiano rispettivamente. Le espressioni figurate contenute nei discorsi originali sono state allineate e messe a confronto con le versioni fornite dagli interpreti, con il duplice obiettivo di a) capire quali causano maggiori problemi agli interpreti e b) analizzare le strategie di interpretazione applicate da professionisti quali quelli della Direzione Generale Interpretazione (DG SCIC) della Commissione europea nell’interpretare metafore. Il progetto prevede anche la somministrazione di un questionario agli interpreti delle cabine spagnola e italiana del DG SCIC, con l’obiettivo di sondare la loro percezione delle difficoltà che sottendono all’interpretazione del linguaggio figurato, le indicazioni metodologiche ricevute (se del caso) dai loro docenti a tale riguardo e le strategie applicate nella pratica professionale. Infine, l’ultima fase del progetto di ricerca prevede la sperimentazione di una proposta didattica attraverso uno studio caso-controllo svolto su studenti del secondo anno della Laurea Magistrale in Interpretazione delle Scuole Interpreti di Forlì e Trieste. Il gruppo-caso ha ricevuto una formazione specifica sull'interpretazione delle metafore, mentre gruppo-controllo è stato monitorato nella sua evoluzione. L’obiettivo di questa ultima fase di ricerca è quello di valutare, da una parte, l’ “insegnabilità” di strategia per affrontare il linguaggio figurato in interpretazione simultanea, e, dall’altra, l’efficacia dell’unità didattica proposta, sviluppata in base all’analisi svolta su IMITES.

Evolutionary epigenetics of modern human populations

Epigenetic variability is a new mechanism for the study of human microevolution, because it creates both phenotypic diversity within an individual and within population. This mechanism constitutes an important reservoir for adaptation in response to new stimuli and recent studies have demonstrated that selective pressures shape not only the genetic code but also DNA methylation profiles. The aim of this thesis is the study of the role of DNA methylation changes in human adaptive processes, considering the Italian peninsula and macro-geographical areas. A whole-genome analysis of DNA methylation profile across the Italian penisula identified some genes whose methylation levels differ between individuals of different Italian districts (South, Centre and North of Italy). These genes are involved in nitrogen compound metabolism and genes involved in pathogens response. Considering individuals with different macro-geographical origins (individuals of Asians, European and African ancestry) more significant DMRs (differentially methylated regions) were identified and are located in genes involved in glucoronidation, in immune response as well as in cell comunication processes. A "profile" of each ancestry (African, Asian and European) was described. Moreover a deepen analysis of three candidate genes (KRTCAP3, MAD1L and BRSK2) in a cohort of individuals of different countries (Morocco, Nigeria, China and Philippines) living in Bologna, was performed in order to explore genetic and epigenetic diversity. Moreover this thesis have paved the way for the application of DNA methylation for the study of hystorical remains and in particular for the age-estimation of individuals starting from biological samples (such as teeth or blood). Noteworthy, a mathematical model that considered methylation values of DNA extracted from cementum and pulp of living individuals can estimate chronological age with high accuracy (median absolute difference between age estimated from DNA methylation and chronological age was 1.2 years).

Searching and retrieving in content-based repositories of formal mathematical knowledge

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.