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.

Integrating mobile devices into a scalable Grid Computing architecture designed to perform distributed computations

The idea of Grid Computing originated in the nineties and found its concrete applications in contexts like the SETI@home project where a lot of computers (offered by volunteers) cooperated, performing distributed computations, inside the Grid environment analyzing radio signals trying to find extraterrestrial life. The Grid was composed of traditional personal computers but, with the emergence of the first mobile devices like Personal Digital Assistants (PDAs), researchers started theorizing the inclusion of mobile devices into Grid Computing; although impressive theoretical work was done, the idea was discarded due to the limitations (mainly technological) of mobile devices available at the time. Decades have passed, and now mobile devices are extremely more performant and numerous than before, leaving a great amount of resources available on mobile devices, such as smartphones and tablets, untapped. Here we propose a solution for performing distributed computations over a Grid Computing environment that utilizes both desktop and mobile devices, exploiting the resources from day-to-day mobile users that alternatively would end up unused. The work starts with an introduction on what Grid Computing is, the evolution of mobile devices, the idea of integrating such devices into the Grid and how to convince device owners to participate in the Grid. Then, the tone becomes more technical, starting with an explanation on how Grid Computing actually works, followed by the technical challenges of integrating mobile devices into the Grid. Next, the model, which constitutes the solution offered by this study, is explained, followed by a chapter regarding the realization of a prototype that proves the feasibility of distributed computations over a Grid composed by both mobile and desktop devices. To conclude future developments and ideas to improve this project are presented.

Micro-FEM models based on micro-CT reconstructions for the in vitro characterization of the elastic properties of trabecular bone tissue.

This master’s thesis describes the research done at the Medical Technology Laboratory (LTM) of the Rizzoli Orthopedic Institute (IOR, Bologna, Italy), which focused on the characterization of the elastic properties of the trabecular bone tissue, starting from october 2012 to present. The approach uses computed microtomography to characterize the architecture of trabecular bone specimens. With the information obtained from the scanner, specimen-specific models of trabecular bone are generated for the solution with the Finite Element Method (FEM). Along with the FEM modelling, mechanical tests are performed over the same reconstructed bone portions. From the linear-elastic stage of mechanical tests presented by experimental results, it is possible to estimate the mechanical properties of the trabecular bone tissue. After a brief introduction on the biomechanics of the trabecular bone (chapter 1) and on the characterization of the mechanics of its tissue using FEM models (chapter 2), the reliability analysis of an experimental procedure is explained (chapter 3), based on the high-scalable numerical solver ParFE. In chapter 4, the sensitivity analyses on two different parameters for micro-FEM model’s reconstruction are presented. Once the reliability of the modeling strategy has been shown, a recent layout for experimental test, developed in LTM, is presented (chapter 5). Moreover, the results of the application of the new layout are discussed, with a stress on the difficulties connected to it and observed during the tests. Finally, a prototype experimental layout for the measure of deformations in trabecular bone specimens is presented (chapter 6). This procedure is based on the Digital Image Correlation method and is currently under development in LTM.

Magnetic actuators for nanosatellite attitude control

The research and the activities presented in the following thesis report have been led at the California Polytechnic State University (US) under the supervision of Prof. Jordi Puig Suari. The objective of the research has been the study of magnetic actuators for nanosatellite attitude control, called magnetorquer. Theese actuators are generally divided in three different kinds: air core torquer, embedded coil and torquerod. In a first phase of the activity, each technology has been analyzed, defining advantages and disadvantages, determining manufacturing procedures and creating mathematical model and designing equation. Dimensioning tools have been then implemented in numerical software to create an instrument that permits to determine the optimal configuration for defined requirements and constraints. In a second phase of the activities the models created have been validated exploiting prototypes and proper instruments for measurements. The instruments and the material exploited for experiments and prototyping have been provided by the PolySat and CubeSat laboratories. The results obtained led to the definition of a complete designing tool and procedure for nanosatellite magnetic actuators, introducing a cost analysis for each kind of solution. The models and the tools have been maintained fully parametric in order to offer a universal re-scalable instrument for satellite of different dimension class.

Synthesis, characterization and formulation of a cathode active material: Copper nitroprusside

Nowadays, rechargeable Li-ion batteries play an important role in portable consumer devices. Formulation of such batteries is improvable by researching new cathodic materials that present higher performances of cyclability and negligible efficiency loss over cycles. Goal of this work was to investigate a new cathodic material, copper nitroprusside, which presents a porous 3D framework. Synthesis was carried out by a low-cost and scalable co-precipitation method. Subsequently, the product was characterized by means of different techniques, such as TGA, XRF, CHN elemental analysis, XRD, Mössbauer spectroscopy and cyclic voltammetry. Electrochemical tests were finally performed both in coin cells and by using in situ cells: on one hand, coin cells allowed different formulations to be easily tested, on the other operando cycling led a deeper insight to insertion process and both chemical and physical changes. Results of several tests highlighted a non-reversible electrochemical behavior of the material and a rapid capacity fading over time. Moreover, operando techniques report that amorphisation occurs during the discharge.

Design and Implementation of a Cloud Infrastructure for Distributed Scientific Calculation

Cloud computing enables independent end users and applications to share data and pooled resources, possibly located in geographically distributed Data Centers, in a fully transparent way. This need is particularly felt by scientific applications to exploit distributed resources in efficient and scalable way for the processing of big amount of data. This paper proposes an open so- lution to deploy a Platform as a service (PaaS) over a set of multi- site data centers by applying open source virtualization tools to facilitate operation among virtual machines while optimizing the usage of distributed resources. An experimental testbed is set up in Openstack environment to obtain evaluations with different types of TCP sample connections to demonstrate the functionality of the proposed solution and to obtain throughput measurements in relation to relevant design parameters.

Sviluppo e applicazione di un reattore di prepolimerizzazione in continuo

Presso lo stabilimento DOW di Correggio (che è una system house per la produzione di prepolimeri poliuretanici) la ricerca è attualmente incentrata sullo sviluppo su scala industriale di un reattore pilota continuo di tipo plug flow che andrà a sostituire i tradizionali processi BATCH, con un guadagno in termini di sicurezza e costi di investimento. In particolare, il progetto prevedeva di sperimentare le “ricette” attuali di prepolimeri per applicazioni nel settore della calzatura per valutarne l’equivalenza con quelli fatti in BATCH, e di calibrare un modello del reattore pilota che permetta di prevedere le prestazioni del reattore e supporti lo scale-up attraverso la raccolta di dati sperimentali (profili di temperatura, tempi di residenza, titolo di NCO del prodotto, ecc...). Alla conclusione del progetto è possibile affermare che la tecnologia è robusta, scalabile e rispetto ai sistemi di produzione attuale presenta una maggiore produttività, sicurezza e minori costi di investimento. At the DOW plant in Correggio, which is a system house for the production of polyurethane prepolymers, the research is currently focused on the develompment and application of a plug flow type continuos reactor that will replace the traditional BATCH processes, with advantages in terms of process safety an investment costs. In particular, the project aims were to test in the pilot plant the prepolymer receipts for footwear application, find out if the result products were similar or better than the ones made with BATCH reactor and harvest experimental data (such as temperatures profiles, reaction time, residual NCO value, etc...) in order to calibrate a model that will support the scale-up to the industrial plant. Now that the project is ended, it is possibile to assert that this tecnology is reliable, scalable, safer and cheaper than the old processes.

A membrane process to extract Citrus Lemon derived extracellular vesicles

Extra cellular vesicles are membrane bound and lipid based nano particles having the size range of 30 to 1000 nm released by a plethora of cells. Their prime function is cellular communication but in the recent studies, the potential of these vesicles to maintain physiological and pathological processes as well as their nano-sized constituents opened doors to its applications in therapeutics, and diagnostics of variety of diseases such as cancer. Their main constituents include lipids, proteins, and RNAs. They are categorized into subtypes such as exosomes, micro-vesicles and apoptotic bodies In recent studies, extracellular vesicles that are derived from plants are gaining high regard due to their variety of advantages such as safety, non-toxicity, and high availability which promotes large scale production. EVs are isolated from mammalian and plant cells using multitude of techniques such as Ultracentrifugation, SEC, Precipitation and so on. Due to the variety in the sources as well as shortcomings arising from the isolation method, a scalable and inexpensive EV isolation method is yet to be designed. This study focusses on isolation of EVs from citrus lemon juice through diafiltration. Lemon is a promising source due to its biological properties to act as antioxidant, anticancer, and anti-inflammatory agents. Lemon derived vesicles was proven to have several proteins analogous to mammalian vesicles. A diafiltration could be carried out for successful removal of impurities and it is a scalable, continuous technique with potentially lower process times. The concentration of purified product and impurities are analysed using Size Exclusion Chromatography in analytical mode. It is also considered imperative to compare the results from diafiltration with gold standard UC. BCA is proposed to evaluate total protein content and DLS for size measurements. Finally, the ideal mode of storage of EVs to protect its internals and its structure is analysed with storage tests.

A Distributed Support based on Kubeflow to Ease MLOps of Federated Learning Algorithms

The main objective of my thesis work is to exploit the Google native and open-source platform Kubeflow, specifically using Kubeflow pipelines, to execute a Federated Learning scalable ML process in a 5G-like and simplified test architecture hosting a Kubernetes cluster and apply the largely adopted FedAVG algorithm and FedProx its optimization empowered by the ML platform ‘s abilities to ease the development and production cycle of this specific FL process. FL algorithms are more are and more promising and adopted both in Cloud application development and 5G communication enhancement through data coming from the monitoring of the underlying telco infrastructure and execution of training and data aggregation at edge nodes to optimize the global model of the algorithm ( that could be used for example for resource provisioning to reach an agreed QoS for the underlying network slice) and after a study and a research over the available papers and scientific articles related to FL with the help of the CTTC that suggests me to study and use Kubeflow to bear the algorithm we found out that this approach for the whole FL cycle deployment was not documented and may be interesting to investigate more in depth. This study may lead to prove the efficiency of the Kubeflow platform itself for this need of development of new FL algorithms that will support new Applications and especially test the FedAVG algorithm performances in a simulated client to cloud communication using a MNIST dataset for FL as benchmark.

Energy consumption of parallel algorithms for solving linear systems on HPC architecture

Modern High-Performance Computing HPC systems are gradually increasing in size and complexity due to the correspondent demand of larger simulations requiring more complicated tasks and higher accuracy. However, as side effects of the Dennard’s scaling approaching its ultimate power limit, the efficiency of software plays also an important role in increasing the overall performance of a computation. Tools to measure application performance in these increasingly complex environments provide insights into the intricate ways in which software and hardware interact. The monitoring of the power consumption in order to save energy is possible through processors interfaces like Intel Running Average Power Limit RAPL. Given the low level of these interfaces, they are often paired with an application-level tool like Performance Application Programming Interface PAPI. Since several problems in many heterogeneous fields can be represented as a complex linear system, an optimized and scalable linear system solver algorithm can decrease significantly the time spent to compute its resolution. One of the most widely used algorithms deployed for the resolution of large simulation is the Gaussian Elimination, which has its most popular implementation for HPC systems in the Scalable Linear Algebra PACKage ScaLAPACK library. However, another relevant algorithm, which is increasing in popularity in the academic field, is the Inhibition Method. This thesis compares the energy consumption of the Inhibition Method and Gaussian Elimination from ScaLAPACK to profile their execution during the resolution of linear systems above the HPC architecture offered by CINECA. Moreover, it also collates the energy and power values for different ranks, nodes, and sockets configurations. The monitoring tools employed to track the energy consumption of these algorithms are PAPI and RAPL, that will be integrated with the parallel execution of the algorithms managed with the Message Passing Interface MPI.