Distributed allocation and scheduling of applications on HPC platforms

High Performance Computing e una tecnologia usata dai cluster computazionali per creare sistemi di elaborazione che sono in grado di fornire servizi molto piu potenti rispetto ai computer tradizionali. Di conseguenza la tecnologia HPC e diventata un fattore determinante nella competizione industriale e nella ricerca. I sistemi HPC continuano a crescere in termini di nodi e core. Le previsioni indicano che il numero dei nodi arrivera a un milione a breve. Questo tipo di architettura presenta anche dei costi molto alti in termini del consumo delle risorse, che diventano insostenibili per il mercato industriale. Un scheduler centralizzato non e in grado di gestire un numero di risorse cosi alto, mantenendo un tempo di risposta ragionevole. In questa tesi viene presentato un modello di scheduling distribuito che si basa sulla programmazione a vincoli e che modella il problema dello scheduling grazie a una serie di vincoli temporali e vincoli sulle risorse che devono essere soddisfatti. Lo scheduler cerca di ottimizzare le performance delle risorse e tende ad avvicinarsi a un profilo di consumo desiderato, considerato ottimale. Vengono analizzati vari modelli diversi e ognuno di questi viene testato in vari ambienti.

Implementation of a non ground meta interpreter for defeasible logic

Human reasoning is a fascinating and complex cognitive process that can be applied in different research areas such as philosophy, psychology, laws and financial. Unfortunately, developing supporting software (to those different areas) able to cope such as complex reasoning it’s difficult and requires a suitable logic abstract formalism. In this thesis we aim to develop a program, that has the job to evaluate a theory (a set of rules) w.r.t. a Goal, and provide some results such as “The Goal is derivable from the KB5 (of the theory)”. In order to achieve this goal we need to analyse different logics and choose the one that best meets our needs. In logic, usually, we try to determine if a given conclusion is logically implied by a set of assumptions T (theory). However, when we deal with programming logic we need an efficient algorithm in order to find such implications. In this work we use a logic rather similar to human logic. Indeed, human reasoning requires an extension of the first order logic able to reach a conclusion depending on not definitely true6 premises belonging to a incomplete set of knowledge. Thus, we implemented a defeasible logic7 framework able to manipulate defeasible rules. Defeasible logic is a non-monotonic logic designed for efficient defeasible reasoning by Nute (see Chapter 2). Those kind of applications are useful in laws area especially if they offer an implementation of an argumentation framework that provides a formal modelling of game. Roughly speaking, let the theory is the set of laws, a keyclaim is the conclusion that one of the party wants to prove (and the other one wants to defeat) and adding dynamic assertion of rules, namely, facts putted forward by the parties, then, we can play an argumentative challenge between two players and decide if the conclusion is provable or not depending on the different strategies performed by the players. Implementing a game model requires one more meta-interpreter able to evaluate the defeasible logic framework; indeed, according to Göedel theorem (see on page 127), we cannot evaluate the meaning of a language using the tools provided by the language itself, but we need a meta-language able to manipulate the object language8. Thus, rather than a simple meta-interpreter, we propose a Meta-level containing different Meta-evaluators. The former has been explained above, the second one is needed to perform the game model, and the last one will be used to change game execution and tree derivation strategies.

Characterization of integrated Bragg gratings in silicon-on-insulator

Silicon-on-insulator (SOI) is rapidly emerging as a very promising material platform for integrated photonics. As it combines the potential for optoelectronic integration with the low-cost and large volume manufacturing capabilities and they are already accumulate a huge amount of applications in areas like sensing, quantum optics, optical telecommunications and metrology. One of the main limitations of current technology is that waveguide propagation losses are still much higher than in standard glass-based platform because of many reasons such as bends, surface roughness and the very strong optical confinement provided by SOI. Such high loss prevents the fabrication of efficient optical resonators and complex devices severely limiting the current potential of the SOI platform. The project in the first part deals with the simple waveguides loss problem and trying to link that with the polarization problem and the loss based on Fabry-Perot Technique. The second part of the thesis deals with the Bragg Grating characterization from again the point of view of the polarization effect which leads to a better stop-band use filters. To a better comprehension a brief review on the basics of the SOI and the integrated Bragg grating ends up with the fabrication techniques and some of its applications will be presented in both parts, until the end of both the third and the fourth chapters to some results which hopefully make its precedent explanations easier to deal with.

Scalable and Seamless Discovery and Selection of Services in Mobile Cloud Computing

Mobile devices are now capable of supporting a wide range of applications, many of which demand an ever increasing computational power. To this end, mobile cloud computing (MCC) has been proposed to address the limited computation power, memory, storage, and energy of such devices. An important challenge in MCC is to guarantee seamless discovery of services. To this end, this thesis proposes an architecture that provides user-transparent and low-latency service discovery, as well as automated service selection. Experimental results on a real cloud computing testbed demonstrated that the proposed work outperforms state of-the-art approaches by achieving extremely low discovery delay.

Studying the alignment of 5-cyanobiphenyl on a polydimethylsiloxane surface through Molecular Dynamics simulations

For 40 years, at the University of Bologna, a group of researchers coordinated by professor Claudio Zannoni has been studying liquid crystals by employing computational techniques. They have developed effective models of these interesting, and still far from being completely understood, systems. They were able to reproduce with simulations important features of some liquid crystal molecules, such as transition temperature. Then they focused their attention on the interactions that these molecules have with different kinds of surface, and how these interactions affect the alignment of liquid crystals. The group studied the behaviour of liquid crystals in contact with different kinds of surfaces, from silica, either amorphous and crystalline, to organic self assembled monolayers (SAMs) and even some common polymers, such as polymethylmethacrylate (PMMA) and polystyrene (PS). Anyway, a library of typical surfaces is still far from being complete, and a lot of work must be done to investigate the cases which have not been analyzed yet. A hole that must be filled is represented by polydimethylsiloxane (PDMS), a polymer on which the interest of industry has enormously grown up in the last years, thanks to its peculiar features, allowing it to be employed in many fields of applications. It has been observed experimentally that PDMS causes 4-cyano-4’-pentylbiphenyl (well known as 5CB), one of the most common liquid crystal molecules, to align homeotropically (i.e. perpendicular) with respect to a surface made of this polymer. Even though some hypothesis have been presented to rationalize the effect, a clear explanation of this phenomenon has not been given yet. This dissertation shows the work I did during my internship in the group of professor Zannoni. The challenge that I had to tackle was to investigate, via Molecular Dynamics (MD) simulations, the reasons of 5CB homeotropic alignment on a PDMS surface, as the group had previously done for other surfaces.

Study of the second hyperpolarizability of poly(3-alkylthiophenes) using the third-harmonic scattering technique

During the past years, the considerable need in the domain of communications for more potent photonic devices has focused the research activities into the nonlinear optical (NLO) materials which can be used for modern optical switches. In this regard, a lot of research activities are focused on the organic materials and conjugated polymers which offer more advantages compared to the inorganic ones. On this matter, poly(3-alkylthiophene) (P3AT), an organic conjugated polymer, can be investigated as potential optical material with in particular the focus on the NLO properties such as the first- and second-hyperpolarizability, β and γ respectively. The activities carried out at the Laboratory of Polymer Synthesis of the KU Leuven, during the master's thesis work, focused on the study of conjugated polymers in order to evaluate their NLO properties for the future purpose of applications in optical systems. In particular, three series of polythiophenes functionalized with an alkyl side chain in the 3-position were synthesized: poly(3-hexylthiophene) (P3HT), poly[3-(2-ethylhexyl)thiophene] (P3EHT) and random copolymer of the two regio-isomers of P3HT. They were made in order to study the influence of molar mass, branching and regio-irregularity on the γ-value. The Kumada catalyst transfer condensative polymerization (KCTCP) and the Pd(RuPhos)-protocol were used for the polymerizations in order to have control over the molar mass of the growing chain and consequently to obtain well-defined and reproducible materials. The P3AT derivatives obtained were characterized by gel permeation chromatography (GPC), spectroscopic techniques (1H-NMR, UV-Vis) and the γ-value was investigated using the third-harmonic scattering (THS) technique. In particular, the THS technique is useful to investigate the optical behavior of the series of polymers in solution.

Performance Analysis of Topological Quantum Error Correcting Codes

In the last few years there has been a great development of techniques like quantum computers and quantum communication systems, due to their huge potentialities and the growing number of applications. However, physical qubits experience a lot of nonidealities, like measurement errors and decoherence, that generate failures in the quantum computation. This work shows how it is possible to exploit concepts from classical information in order to realize quantum error-correcting codes, adding some redundancy qubits. In particular, the threshold theorem states that it is possible to lower the percentage of failures in the decoding at will, if the physical error rate is below a given accuracy threshold. The focus will be on codes belonging to the family of the topological codes, like toric, planar and XZZX surface codes. Firstly, they will be compared from a theoretical point of view, in order to show their advantages and disadvantages. The algorithms behind the minimum perfect matching decoder, the most popular for such codes, will be presented. The last section will be dedicated to the analysis of the performances of these topological codes with different error channel models, showing interesting results. In particular, while the error correction capability of surface codes decreases in presence of biased errors, XZZX codes own some intrinsic symmetries that allow them to improve their performances if one kind of error occurs more frequently than the others.

An optimal control tool for trajectory generation of autonomous drones

In the recent years, autonomous aerial vehicles gained large popularity in a variety of applications in the field of automation. To accomplish various and challenging tasks the capability of generating trajectories has assumed a key role. As higher performances are sought, traditional, flatness-based trajectory generation schemes present their limitations. In these approaches the highly nonlinear dynamics of the quadrotor is, indeed, neglected. Therefore, strategies based on optimal control principles turn out to be beneficial, since in the trajectory generation process they allow the control unit to best exploit the actual dynamics, and enable the drone to perform quite aggressive maneuvers. This dissertation is then concerned with the development of an optimal control technique to generate trajectories for autonomous drones. The algorithm adopted to this end is a second-order iterative method working directly in continuous-time, which, under proper initialization, guarantees quadratic convergence to a locally optimal trajectory. At each iteration a quadratic approximation of the cost functional is minimized and a decreasing direction is then obtained as a linear-affine control law, after solving a differential Riccati equation. The algorithm has been implemented and its effectiveness has been tested on the vectored-thrust dynamical model of a quadrotor in a realistic simulative setup.

Acquisition of Vibration Signals for Predictive Maintenance: a Practical Case

As predictive maintenance becomes more and more relevant in industrial environment, the possible range of applications for this maintenance strategy grows. The progresses in components technology and their reduction in price, together with the late years' advances in machine learning and in computational power, are making the implementation of predictive maintenance possible in plants where it would have previously been unreasonably costly. This is leading major pharmaceutical industries to explore the possibility of the application of condition monitoring systems on progressively less and less critical equipment. The focus of this thesis is on the implementation of a system to gather vibrational data from the motors installed in a pre-existing machine using off-the-shelf components. The final goal for the system is to provide the necessary vibration data, in the form of frequency spectra, to a machine learning system developed by IMA Digital, which will be leveraging such data to predict possible upcoming faults and to give the final client all the information necessary to plan maintenance activity according to the estimated machine condition.

Application of active and transfer learning to wound image segmentation

Wound management is a fundamental task in standard clinical practice. Automated solutions already exist for humans, but there is a lack of applications on wound management for pets. The importance of a precise and efficient wound assessment is helpful to improve diagnosis and to increase the effectiveness of treatment plans for the chronic wounds. The goal of the research was to propose an automated pipeline capable of segmenting natural light-reflected wound images of animals. Two datasets composed by light-reflected images were used in this work: Deepskin dataset, 1564 human wound images obtained during routine dermatological exams, with 145 manual annotated images; Petwound dataset, a set of 290 wound photos of dogs and cats with 0 annotated images. Two implementations of U-Net Convolutioal Neural Network model were proposed for the automated segmentation. Active Semi-Supervised Learning techniques were applied for human-wound images to perform segmentation from 10% of annotated images. Then the same models were trained, via Transfer Learning, adopting an Active Semi- upervised Learning to unlabelled animal-wound images. The combination of the two training strategies proved their effectiveness in generating large amounts of annotated samples (94% of Deepskin, 80% of PetWound) with the minimal human intervention. The correctness of automated segmentation were evaluated by clinical experts at each round of training thus we can assert that the results obtained in this thesis stands as a reliable solution to perform a correct wound image segmentation. The use of Transfer Learning and Active Semi-Supervied Learning allows to minimize labelling effort from clinicians, even requiring no starting manual annotation at all. Moreover the performances of the model with limited number of parameters suggest the implementation of smartphone-based application to this topic, helping the future standardization of light-reflected images as acknowledge medical images.

Stabilizzazione del Blu di Prussia con film politiofenici depositati per via elettrochimica

Hydrogen peroxide (H2O2) is a powerful oxidant which is commonly used in a wide range of applications in the industrial field. Several methods for the quantification of H2O2 have been developed. Among them, electrochemical methods exploit the ability of some hexacyanoferrates (such as Prussian Blue) to detect H2O2 at potentials close to 0.0 V (vs. SCE) avoiding the occurrence of secondary reactions, which are likely to run at large overpotentials. This electrocatalytic behaviour makes hexacyanoferrates excellent redox mediators. When deposited in the form of thin films on the electrode surfaces, they can be employed in the fabrication of sensors and biosensors, normally operated in solutions at pH values close to physiological ones. As hexacyanoferrates show limited stability in not strongly acidic solutions, it is necessary to improve the configuration of the modified electrodes to increase the stability of the films. In this thesis work, organic conducting polymers were used to fabricate composite films with Prussian Blue (PB) to be electro-deposited on Pt surfaces, in order to increase their pH stability. Different electrode configurations and different methods of synthesis of both components were tested, and for each one the achievement of a possible increase in the operational stability of Prussian Blue was verified. Good results were obtained for the polymer 3,3''-didodecyl-2,2':5',2''-terthiophene (poly(3,3''-DDTT)), whose presence created a favourable microenvironment for the electrodeposition of Prussian Blue. The electrochemical behaviour of the modified electrodes was studied in both aqueous and organic solutions. Poly(3,3''-DDTT) showed no response in aqueous solution in the potential range where PB is electroactive, thus in buffered aqueous solution is was possible to characterize the composite material, focusing only on the redox behaviour of PB. A combined effect of anion and cation of the supporting electrolyte was noticed. The response of Pt electrodes modified with films of the PB /poly(3,3''-DDTT) composite was evaluated for the determination of H2O2. The performance of such films was found better than that of the PB alone. It can be concluded that poly(3,3''-DDTT) plays a key role in the stabilization of Prussian Blue causing also a wider linearity range for the electrocatalytic response to H2O2.

Studio e sviluppo di tecniche per la produzione di nanocompositi a matrice di alluminio

Foundry aluminum alloys play a fundamental role in several industrial fields, as they are employed in the production of several components in a wide range of applications. Moreover, these alloys can be employed as matrix for the development of Metal Matrix Composites (MMC), whose reinforcing phases may have different composition, shape and dimension. Ceramic particle reinforced MMCs are particular interesting due to their isotropic properties and their high temperature resistance. For this kind of composites, usually, decreasing the size of the reinforcing phase leads to the increase of mechanical properties. For this reason, in the last 30 years, the research has developed micro-reinforced composites at first, characterized by low ductility, and more recently nano-reinforced ones (the so called metal matrix nanocomposite, MMNCs). The nanocomposites can be obtained through several production routes: they can be divided in in-situ techniques, where the reinforcing phase is generated during the composite production through appropriate chemical reactions, and ex situ techniques, where ceramic dispersoids are added to the matrix once already formed. The enhancement in mechanical properties of MMNCs is proved by several studies; nevertheless, it is necessary to address some issues related to each processing route, as the control of process parameters and the effort to obtain an effective dispersion of the nanoparticles in the matrix, which sometimes actually restrict the use of these materials at industrial level. In this work of thesis, a feasibility study and implementation of production processes for Aluminum and AlSi7Mg based-MMNCs was conducted. The attention was focused on the in-situ process of gas bubbling, with the aim to obtain an aluminum oxide reinforcing phase, generated by the chemical reaction between the molten matrix and industrial dry air injected in the melt. Moreover, for what concerns the ex-situ techniques, stir casting process was studied and applied to introduce alumina nanoparticles in the same matrix alloys. The obtained samples were characterized through optical and electronic microscopy, then by micro-hardness tests, in order to evaluate possible improvements in mechanical properties of the materials.

Differential forms on Carnot groups

The main goal of this thesis is to understand and link together some of the early works by Michel Rumin and Pierre Julg. The work is centered around the so-called Rumin complex, which is a construction in subRiemannian geometry. A Carnot manifold is a manifold endowed with a horizontal distribution. If further a metric is given, one gets a subRiemannian manifold. Such data arise in different contexts, such as: - formulation of the second principle of thermodynamics; - optimal control; - propagation of singularities for sums of squares of vector fields; - real hypersurfaces in complex manifolds; - ideal boundaries of rank one symmetric spaces; - asymptotic geometry of nilpotent groups; - modelization of human vision. Differential forms on a Carnot manifold have weights, which produces a filtered complex. In view of applications to nilpotent groups, Rumin has defined a substitute for the de Rham complex, adapted to this filtration. The presence of a filtered complex also suggests the use of the formal machinery of spectral sequences in the study of cohomology. The goal was indeed to understand the link between Rumin's operator and the differentials which appear in the various spectral sequences we have worked with: - the weight spectral sequence; - a special spectral sequence introduced by Julg and called by him Forman's spectral sequence; - Forman's spectral sequence (which turns out to be unrelated to the previous one). We will see that in general Rumin's operator depends on choices. However, in some special cases, it does not because it has an alternative interpretation as a differential in a natural spectral sequence. After defining Carnot groups and analysing their main properties, we will introduce the concept of weights of forms which will produce a splitting on the exterior differential operator d. We shall see how the Rumin complex arises from this splitting and proceed to carry out the complete computations in some key examples. From the third chapter onwards we will focus on Julg's paper, describing his new filtration and its relationship with the weight spectral sequence. We will study the connection between the spectral sequences and Rumin's complex in the n-dimensional Heisenberg group and the 7-dimensional quaternionic Heisenberg group and then generalize the result to Carnot groups using the weight filtration. Finally, we shall explain why Julg required the independence of choices in some special Rumin operators, introducing the Szego map and describing its main properties.