Search

The aim of this work is primarily the adaptation of the communication structure and its environment, in which the on-board services of the Drone Engineering Ecosystem (DEE) are located. The objective is to facilitate the installation and use of these services within the drone, also improving their efficiency while being executed, through the use of a technology known as Docker, commonly employed in many production environments. To achieve this goal, the corresponding Dockerfile will be developed for each required image, with the intention of creating the set of containers that will form the Docker-based ecosystem. All these containers will be grouped according to the configuration established in a file named docker-compose.yml, based on the Docker Compose tool, used for the management of several containers. In addition to this development, this project also focuses on the creation of an air backend, located on the drone's Raspberry Pi, and a ground backend, located on the Classpip server, available as part of another collaborative project within the EETAC. The goal of the development of these two storage points, is to save in them information associated with flight plans and executed flights, in relation with Dashboard application and the Flutter developed mobile application, two different components within the drone ecosystem. Once these two new storage components are implemented within the ecosystem, the necessary modifications and contributions will be made to enable these mentioned applications, both the Dashboard and the one developed in Flutter, to create flight plans, capture images and videos, and subsequently visualize all these multimedia files within the applications themselves, for a proper analysis of the obtained results. With the intention of verifying the set of contributions made to the ecosystem, the relevant tests will be conducted on real flights at the DroneLab, to assess the effectiveness of the implemented modifications, communications, and st

Títol alternatiu emprat a la intranet de l'EPSEVG: "Tecnologia blockchain per l’autenticació de dispositius “un-trusted” IoT ", Aquest treball final de grau ens mostra una manera per solucionar un aspecte negatiu dels dispositius IoT com és poder confiar en la integritat de les dades proporcionades. Aquest problema el podrem solucionar aplicant un mètode d’ autenticació de dispositius d’una forma segura i original fent ús d’eines de codi obert com el marc de treball de cadena de blocs Hyperledger Fabric, la qual servirà per emmagatzemar les dades per futures autenticacions de dispositius. S’empraran proves zkp més específicament zk-snark, utilitzant la llibreria libsnark, per la autenticació de dispositius de forma segura fent ús del protocol HTTPS per la comunicació entre el dispositiu i l’aplicació creada per comunicar-se amb la cadena de blocs. Per poder construir un escenari realista, hi ha una xarxa Docker on es troben els components de la cadena de blocs simbolitzant així un entorn empresarial. Per altra banda hi ha la simbolització d’un dispositiu IoT que es comunica amb l’aplicació per realitzar peticions que seran ateses. Si el dispositiu està registrat al sistema serà autenticat i podrà accedir de manera segura a la informació. La solució aportada és robusta i segura i pot ser aplicada en múltiples àmbits on es requereix una autenticació fiable., Este trabajo final de grado nos muestra una forma para solucionar un aspecto negativo de los dispositivos IoT como es poder confiar en la integridad de los datos proporcionados. Este problema se podrá solucionar aplicando un método de autenticación de dispositivos de una forma segura y original haciendo uso de herramientas de código abierto como el marco de trabajo de cadena de bloques Hyperledger Fabric, que servirá para almacenar los datos para futuras autenticaciones de dispositives. Se utilizarán pruebas de conocimiento cero (zkp) más específicamente zk-snark, utilizando la librería libsnark, para la autenticación de dispositivos de forma segura, haciendo uso del protocolo HTTPS para la comunicación entre el dispositivo y la aplicación creada para comunicarse con la cadena de bloques. Para poder construir un escenario realista, existe una red Docker donde se encuentran los componentes de la cadena de bloques, simbolizando de esta manera un entorno empresarial. Por otra parte, existe la simulación de un dispositivo IoT que se comunica con la aplicación para realizar peticiones que serán atendidas. Si el dispositivo está registrado en el sistema, será autenticado y podrá acceder de forma segura a la información. La solución aportada es robusta y segura y puede aplicarse en muchos ámbitos distintos donde se requiera una autenticación fiable., My end of degree thesis aims at showing a way to solve a negative aspect of IoT devices which is the need to be able to trust the integrity of the data provided. We are able to solve this problem by applying a method of device authentication in a secure and original way while using open-source tools such as the Hyperledger Fabric blockchain framework, which is used to store de data for future authentications of devices. ZKP tests will be used, specifically ZK-snark, and the libsnark library, to authenticate devices safely using the HTTPS protocol for the communication between the device and the application created to communicate with the blockchain. To build a realistic scenario, there is a Cocker network where the blockchain components are located, symbolizing in this way an enterprise environment. On the other hand, there is the symbolization of a IoT device that communicates with the application to make requests that will be answered. If the device is registered in the system, it will be authenticated and will have a secure access to the information requested. The solution provided is robust and secure and can be applied in various areas where reliable authentication is required.

Nowadays, the space industry is experiencing exponential growth, fueled by private companies and the conception of new and ambitious challenges. In recent years, the cost of reaching space has been drastically reduced, and to a certain extent, reusable rockets are to blame for that cost reduction. Unfortunately, and it has always been so, the space industry has always required a lot of resources and funding, and even with the drastic cost reduction with the incorporation of this new type of rockets, the development of technologies by students and researchers from universities around the world is still very difficult. To address this problem, this thesis will attempt, by reducing the cost to a maximum, to develop a system where it can study the dynamics of reusable rockets and be able to develop technologies around them. To do this, an unmanned aircraft system (UAS) will be designed from scratch with the main objective of imitating the behavior of this type of rocket while trying to reuse as many components already used by the university as possible, helping to reduce costs, and also helping to reduce the environmental impact of the project itself. Through the different chapters of this work, it can be seen, from the conception of the initial requirements and selection of the different components, to the 3D design and subsequent manufacturing of the system itself, which will be calibrated and tested in different test flights, taking advantage of the university's facilities. Ultimately, this work aims to serve as a useful tool for future students and researchers to deepen their knowledge of the attitude control of this type of rockets, and furthermore, to develop new technologies around the system that is intended to be designed and manufactured in this thesis.

In recent years, a growing awareness of the importance of a healthy lifestyle has fueled a surge in the demand for personal trainers. The ongoing COVID-19 pandemic has further emphasized the importance of accessible fitness solutions. With lockdowns and restrictions limiting traditional gym access, individuals have sought alternative means to maintain their health and well-being. Fitness applications, offering at-home workouts and virtual coaching, have become essential tools for many during these challenging times. The pandemic has, in essence, served as a propulsor, encouraging a broader demographic to engage in regular workouts and adopt a healthier lifestyle. Recognizing these challenges, the development of a fitness application becomes pivotal in bridging fitness enthusiasts a professional guidance along workouts. Many individuals aspiring to go to the gym lack the necessary knowledge or financial means to hire a personal trainer. A fitness application, particularly one developed in Swift with embedded machine-learning capabilities, can address these handicaps effectively acting as a virtual personal trainer, thus, offering personal guidance when doing specific exercises. Moreover, a fitness application can go beyond the role of a traditional personal trainer and serve as a social network, fostering a sense of community among users. This social aspect encourages motivation, friendly competition, and the exchange of tips and experiences. Additionally, the application can function as a comprehensive registry, allowing users to track their fitness journey, monitor achievements, and set new goals. This multifaceted approach enhances the overall user experience and provides a holistic solution to the challenges faced by fitness enthusiasts. This document aims to explore and highlight how a fitness application developed in Swift with embedded machine-learning capabilities can effectively serve as a personal trainer, a social network, and a registry app. By providing

El desarrollo de este proyecto se fundamentará en la creación de herramientas destinadas a la obtención de un conjunto de datos para realizar un análisis termográfico de algunos edificios situados en el campus Nord de la UPC. Este análisis se basará en la información recopilada por una red de sensores encargados de medir la calidad del aire, que se instalarán como punto de partida del proyecto. Además de estos sensores, se llevará a cabo la implementación de estaciones meteorológicas y sensores de temperatura en la superficie de los edificios. El conjunto de datos se complementa con un grupo de imágenes capturadas por un dron. Las herramientas diseñadas abarcan el desarrollo de nodos de monitorización y la aplicación de técnicas avanzadas para el análisis de los datos recabados. Este enfoque integral busca proporcionar una visión detallada y precisa de las condiciones ambientales y térmicas, contribuyendo así a un análisis exhaustivo de los edificios en cuestión.

The extensive use of HPC infrastructures and frameworks for running data-intensive applications has led to a growing interest in data partitioning techniques and strategies. In fact, application performance can be heavily affected by how data are partitioned, which in turn depends on the selected size for data blocks, i.e. the block size. Therefore, finding an effective partitioning, i.e. a suitable block size, is a key strategy to speed-up parallel data-intensive applications and increase scalability. This paper describes a methodology, namely BLEST-ML (BLock size ESTimation through Machine Learning), for block size estimation that relies on supervised machine learning techniques. The proposed methodology was evaluated by designing an implementation tailored to dislib, a distributed computing library highly focused on machine learning algorithms built on top of the PyCOMPSs framework. We assessed the effectiveness of the provided implementation through an extensive experimental evaluation considering different algorithms from dislib, datasets, and infrastructures, including the MareNostrum 4 supercomputer. The results we obtained show the ability of BLEST-ML to efficiently determine a suitable way to split a given dataset, thus providing a proof of its applicability to enable the efficient execution of data-parallel applications in high performance environments., This work has been partially supported by the European Commission through the Horizon 2020 Research and Innovation program and the EuroHPC JU under contract 955558 (eFlows4HPC project) and by MCIN/AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR (PCI2021‑121957 and CEX2021‑ 001148‑S) and by the Spanish Government (PID2019‑107255GB), and Generalitat de Catalunya (contract 2021‑SGR‑00412). We also acknowledge financial support from “National Centre for HPC, Big Data and Quantum Computing”, CN00000013 ‑ CUP H23C22000360005, and from “FAIR ‑ Future Artificial Intelligence Research” project ‑ CUP H23C22000860006., Peer Reviewed, Postprint (published version)

The full deployment of sixth-generation (6G) networks is inextricably connected with a holistic network redesign able to deal with various emerging challenges, such as integration of heterogeneous technologies and devices, as well as support of latency and bandwidth demanding applications. In such a complex environment, resource optimization, and security and privacy enhancement can be quite demanding, due to the vast and diverse data generation endpoints and associated hardware elements. Therefore, efficient data collection mechanisms are needed that can be deployed at any network infrastructure. In this context, the network data analytics function (NWDAF) has already been defined in the fifth-generation (5G) architecture from Release 15 of 3GPP, that can perform data collection from various network functions (NFs). When combined with advanced machine learning (ML) techniques, a full-scale network optimization can be supported, according to traffic demands and service requirements. In addition, the collected data from NWDAF can be used for anomaly detection and thus, security and privacy enhancement. Therefore, the main goal of this paper is to present the current state-of-the-art on the role of the NWDAF towards data collection, resource optimization and security enhancement in next generation broadband networks. Furthermore, various key enabling technologies for data collection and threat mitigation in the 6G framework are identified and categorized, along with advanced ML approaches. Finally, a high level architectural approach is presented and discussed, based on the NWDAF, for efficient data collection and ML model training in large scale heterogeneous environments., This work was supported in part by the HORSE Project funded by the Smart Networks and Services Joint Undertaking (SNS JU) through the European Union’s Horizon Europe Research and Innovation Program under Grant 101096342 (www.horse-6g.eu)., Peer Reviewed, Postprint (published version)

This document outlines the conception, development, and implementation of a system dedicated to managing conflicts among drones (UAVs). Throughout the course of this work, solutions to various problems that may arise in environments with multiple flying aircraft are presented. In response to these issues, two algorithms have been developed, along with a communication structure/protocol between agents, which combined are capable of handling a vast array of possible real-world scenarios. The system proposes the use of dedicated software and hardware through which all drones carrying it can establish vehicle-to-vehicle communication. Through this communication, it is expected that necessary data, including flight plans, can be exchanged, enabling each involved agent to locally initiate the execution of the proposed algorithms. These algorithms will provide a conflict-free 3D solution for the flight plans of all drones in the network. For the calculation of this solution, the use of priorities has been suggested, aligning with real-life situations and providing the system with a means of discerning which of the involved drones should have more or fewer restrictions. The first algorithm involves, given the routes to be followed by all connected agents, determining if these routes meet the requirements to be treated as high-conflict routes. If so, the drones executing them will have their routes altered based on their priority by the "Reciprocal heading obstacle" (RHO) algorithm. Once a solution has been found for this type of high-conflict route, the "Traffic Lights" algorithm is executed. This second algorithm, based on the modified routes by RHO and the defined priorities, calculates at which points the aircraft should wait in the air to yield the right of way to higher-priority traffic. Once these algorithms provide a local solution for each drone, this solution is communicated to the rest of the swarm to determine if each one's solution is compatible with others. Thi, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles

Nowadays, companies collect massive quantities of data to enhance their operations, often at the expense of sharing user sensible information. This data is widely used to train Deep Learning (DL) neural networks to model, classify, or recognize complex data. These activities enable companies to offer an array of services to users, such as precise advertising and optimal location services. This study explores potential solutions for preserving privacy while utilizing DL applications. To address the privacy issue, we develop a privacy-preserving framework specifically designed for fog computing environments. Unlike traditional cloud computing architectures, fog embedded architectures only share a small portion of user data with a nearby fog node, ensuring that the majority of sensitive data remains secure. Within these fog nodes, we incorporate two additional algorithms, namely Generalization and Threshold, to enhance the privacy-preserving capabilities of the framework. The first algorithm, Generalization, introduces a validation dataset within the fog nodes which not only increases the accuracy of the fog-embedded framework but also ensures that user data is preserved. The second algorithm, Threshold, is responsible for protecting user data samples and reducing the amount of information sent to the server. By combining these two algorithms, we are able to provide an additional layer of protection for user privacy while still maintaining the accuracy of the model. We conduct an evaluation to test its effectiveness using two separate datasets. In addition, we analyze them through a Feed Forward Neural Network (FFNN) and compare the results with a traditional centralized architecture to validate the effectiveness of the proposed framework. The results of our evaluation demonstrate that the proposed privacy-preserving framework, when combined with the Generalization and Threshold algorithms, can preserve up to 38.44% of user data. Additionally, we were able to extend th, This work is partially supported by the Spanish Ministry of Science and Innovation, Spain under contracts PID2021-124463OB-IOO and PID2019-107255GB-C22; by the Generalitat de Catalunya, Spain under grant 2021SGR00326; and by the DRAC (IU16-011591), the HORIZON Vitamin-V, Spain(101093062), the HORIZON-AG PHOENI2X, Spain (101070586) and the HORIZON HORSE, Spain (101096342) projects., Peer Reviewed, Postprint (published version)

Large language models (LLMs) have revolutionized the state-of-the-art of many different natural language processing tasks. Although serving LLMs is computationally and memory demanding, the rise of Small Language Models (SLMs) offers new opportunities for resource-constrained users, who now are able to serve small models with cutting-edge performance. In this paper, we present a set of experiments designed to benchmark SLM inference at performance and energy levels. Our analysis provides a new perspective in serving, highlighting that the small memory footprint of SLMs allows for reaching the Pareto-optimal throughput within the resource capacity of a single accelerator. In this regard, we present an initial set of findings demonstrating how model replication can effectively improve resource utilization for serving SLMs., This work has been partially financed by grant agreement EUHORIZON GA.101095717 and by the EU-HORIZON MSCA programme under grant agreement EU-HORIZON MSCA GA.101086248. Also, it has been partially financed by Generalitat de Catalunya (AGAUR) under grant agreement 2021- SGR-00478, and by the Spanish Ministry of Science (MICINN), the Research State Agency (AEI) and European Regional Development Funds (ERDF/FEDER) under grant agreement PID2021-126248OB-I00, MCIN/AEI/ 10.13039/ 501100011033/ FEDER, UE., Peer Reviewed, Postprint (author's final draft)

This document describes all the work involved in creating a code library for the control of drones and their cameras. This library aims to replace the Dronekit library, a public library currently discontinued, which is present in most modules of the Drone Engineering Ecosystem at the School of Telecommunication and Aerospace Engineering of Castelldefels. Specifically, the development process of two new libraries will be described: DroneLink EETAC, which will be oriented towards drone control, and CameraLink EETAC, designed to manage the integrated camera on the drone. Additionally, the construction of three examples of use for these libraries will be included: Dashboard Direct, Dashboard Direct Multiple, and Dashboard Remote. Each will demonstrate different practical applications of the libraries, thereby highlighting their potential and versatility. With Dashboard Direct, the libraries will be used in a local context, through a direct connection to the drone via telemetry radio. This will be the first example developed and will help to verify the correct operation of the libraries. Dashboard Direct Multiple will allow the control of a swarm of up to 10 drones through a single device. This will be the second example developed and will highlight the great modularity and optimization of the libraries through their architecture. The third example will be Dashboard Remote, a desktop application that will allow the user to control the drone remotely through an external broker. In this case, the libraries will run on services integrated into the drone. This example will also cover other interesting topics, such as real-time video transmission using the websocket protocol. The development of DroneLink EETAC aims to lay the foundations for a new technology with which to build applications or services aimed at aerial vehicles. In this way, dependency on the continuity of elements developed by users external to the school will be eliminated.

Aquest treball de final de grau es centra en el desenvolupament d'un sistema automatitzat de generació de newsletters per a la plataforma Argos City, que ofereix dades meteorològiques i hidrològiques als seus usuaris. L'objectiu principal del projecte és millorar l'accessibilitat a les dades meteorològiques, personalitzar el contingut de les newsletters segons les preferències dels usuaris i automatitzar el procés de generació i enviament de les newsletters. El sistema s'ha desenvolupat aprofitant l'arquitectura existent de la plataforma Argos City. S'ha utilitzat .NET (C#) per al backend que gestiona la lògica de l'aplicació i processa les sol·licituds dels usuaris, mentre que el frontend s'ha creat amb Vue.js per oferir una interfície d'usuari interactiva., This final degree project focuses on the development of an automated newsletter generation system for the Argos City platform, which provides meteorological and hydrological data to its users. The main objective of the project is to improve access to meteorological data, personalize the content of newsletters according to user preferences, and automate the process of generating and sending newsletters. The system was developed by leveraging the existing architecture of the Argos City platform. For the backend it was used .NET (C#), which manages application logic and processes user requests, while the frontend was created with Vue.js to offer an interactive user interface.

Bluetooth technology is very common for communication between devices such as mobile phones, headphones, and even watches. Nowadays, it is difficult to find someone who does not have at least one of them. This opens the door to the possibility of detecting these devices using this technology, and with it, the individual carrying them. Over the years, many people have gotten lost in forests or mountainous areas due to disorientation, sudden weather changes, or countless other reasons. A lost and uncommunicated individual represents an emergency that requires quick and effective rescue methods. The methods used today by rescue teams, such as helicopters and drones equipped with cameras, involve visually searching for the individual, which can slow down the process and increase the risk to the lost person. To address this problem, this project, based on the integration of two previous projects, has designed a system capable of measuring Bluetooth signal strength, a parameter called RSSI, in different orientations in space. This system is mounted on a drone, allowing large areas to be covered in a short period of time. Throughout this thesis, we will explore the different hardware components used for detection, how they are connected, and the supports designed to safely mount the system on the drone while minimizing payload. The software and intelligence used to capture and process RSSI data will be detailed. Finally, the project will conclude by presenting the results of flight tests conducted in controlled environments, such as the university facilities, as well as in more realistic environments to observe the system's behavior at greater distances and in wooded areas. Overall, this project aims to combine the conclusions from two previous projects to create a system capable of detecting Bluetooth signal strength and its direction of origin. This opens up new possibilities in the field of locating people using drone technology, Objectius de Desenvolupament Sostenible::7 - Energia Assequible i No Contaminant, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura

One of the main goals of human genetics is to understand the connections between genomic variation and the predisposition to develop a complex disorder. These disease–variant associations are usually studied in a single independent manner, disregarding the possible effect derived from the interaction between genomic variants. In particular, in a background of complex diseases, these interactions can be directly linked to the disorder and may play an important role in disease development. Although their study has been suggested to help complete the understanding of the genetic bases of complex diseases, this still represents a big challenge due to large computing demands. Here, we take advantage of high-performance computing technologies to tackle this problem by using a combination of machine learning methods and statistical approaches. As a result, we created a containerized framework that uses multifactor dimensionality reduction (MDR) to detect pairs of variants associated with type 2 diabetes (T2D). This methodology was tested on the Northwestern University NUgene project cohort using a dataset of 1,883,192 variant pairs with a certain degree of association with T2D. Out of the pairs studied, we identified 104 significant pairs: two of which exhibit a potential functional relationship with T2D. These results place the proposed MDR method as a valid, efficient, and portable solution to study variant interaction in real reduced genomic datasets., This work was partially financed by the European Commission (EU-HORIZON NEARDATA GA.101092644) and by the Universitat Politècnica de Catalunya (45-FPIUPC2018); it was also partially financed by Generalitat de Catalunya (AGAUR) under grant agreement 2021-SGR-00478; it was also partially financed by the Spanish Ministry of Science (MICINN), the Research State Agency (AEI), and European Regional Development Funds (ERDF/FEDER) under grant agreement PID2021-126248OB-I00, MCIN/AEI/10.13039/501100011033/FEDER, UE., Peer Reviewed, Postprint (published version)

Tesi amb menció de Doctorat Internacional, Tesi en modalitat de cotutela: Universitat Politècnica de Catalunya i Università degli studi di Venezia, (English) This thesis aims to investigate the efficacy and applicability of Wireless Backhaul Network (WBN) in two divergent contexts: ultra-dense urban networks for 5G and connectivity solutions for rural or digitally divided areas. WBN offer an adaptable and resilient framework for data transmission, making them an attractive option for the next generation of wireless networks, particularly 5G. In dense urban settings, where the demand for high data rates is pressing, wireless mesh backhauls can serve as a strategic asset in achieving the promised data throughput for 5G networks. These networks inherently require a dense deployment of nodes to deliver on their promise of high-speed, low-latency communication. Therefore, we analyze the role that wireless mesh backhauls can play in such densely populated areas, emphasizing their potential to meet or even exceed 5G’s high data rate expectations. Conversely, in rural or digitally divided areas, the economic feasibility of deploying traditional last-mile copper or fiber-optic networks often proves to be prohibitive. In such cases, WBN, deployed either as Wireless Community Network (WCN) or as part of a Wireless Internet Service Provider (WISP)’s infrastructure, can offer a viable alternative. This technology could bridge the digital divide by providing robust, cost-effective internet connectivity to underserved regions. To substantiate my findings, I leveraged Geographic Information Systems (GIS) technology and GPU-based computational methods together with graph analysis and simulations. GIS technology, leveraging different open datasets such as Digital Surface Models (DSM) and vectorial maps, make a detailed understanding of the geographical landscape possible, crucial for building detailed feasibility models in various environments. GPU-based computational methods accelerate the model analysis, enabling an in-depth evaluation within a reasonable timeframe. As a result I was able to optimize various network aspects, s, (Català) Aquesta tesi té com a objectiu investigar l'eficàcia i l'aplicabilitat de la xarxa de backhaul sense fil (WBN) en dos contextos divergents: xarxes urbanes ultradenses per a 5G i solucions de connectivitat per a zones rurals o dividides digitalment. WBN ofereix un marc adaptable i resistent per a la transmissió de dades, cosa que els converteix en una opció atractiva per a la propera generació de xarxes sense fils, especialment 5G. En entorns urbans densos, on la demanda d'elevades taxes de dades és urgent, els backhauls de malla sense fils poden servir com un actiu estratègic per assolir el rendiment de dades promès per a les xarxes 5G. Aquestes xarxes requereixen inherentment un desplegament dens de nodes per complir la seva promesa de comunicació d'alta velocitat i baixa latència. Per tant, analitzem el paper que poden tenir els backhauls de malla sense fils en zones densament poblades, posant èmfasi en el seu potencial per complir o fins i tot superar les expectatives d'alta velocitat de dades de 5G. Per contra, a les zones rurals o digitalment dividides, la viabilitat econòmica de desplegar xarxes tradicionals de coure o fibra òptica d'última milla sovint resulta prohibitiva. En aquests casos, WBN, desplegat com a xarxa comunitària sense fils (WCN) o com a part de la infraestructura d'un proveïdor de serveis d'Internet sense fils (WISP), pot oferir una alternativa viable. Aquesta tecnologia podria reduir la bretxa digital proporcionant una connectivitat d'Internet robusta i rendible a les regions poc ateses. Per corroborar les meves troballes, vaig aprofitar la tecnologia de sistemes d'informació geogràfica (GIS) i mètodes computacionals basats en GPU juntament amb anàlisis i simulacions de gràfics. La tecnologia GIS, aprofitant diferents conjunts de dades oberts com ara els models de superfície digital (DSM) i els mapes vectorials, fan possible una comprensió detallada del paisatge geogràfic, crucial per construir models de viabilitat detallats en dive, Postprint (published version)

Este trabajo se ha realizado con la empresa HP Inc., y surge de la necesidad de rediseñar el sistema de monitorización de una infraestructura de máquinas en un entorno de integración continua. El objetivo es conseguir una mayor automatización, portabilidad y escalabilidad del sistema, minimizando así la intervención manual y adaptándose mejor a mayores volúmenes de datos y dispositivos. El proyecto define el sistema actual y los problemas que este presenta y, después de un estudio de las diferentes posibles soluciones, detalla la solución propuesta adaptándose a los requerimientos del nuevo sistema. La solución propuesta y posteriormente desarrollada, se basa en la tecnología Zabbix, y consta de un servidor de monitorización que almacena y analiza los datos enviados por los hosts, las máquinas que conforman el sistema y que se quieren monitorizar. Tanto la creación del servidor, como la inserción de los hosts en el sistema, se realiza mediante el uso de Ansible, tecnología que permite la automatización completa de la tarea. La recolección de datos se ha realizado mediante la monitorización de ítems de las máquinas del sistema, como la carga de CPU o el estado de los servicios que ejecutan. Para ello, se han utilizado ítems preestablecidos por Zabbix y también se ha desarrollado un ciclo de desarrollo que permite crear ítems personalizados. Zabbix ofrece una interfaz gráfica para la configuración del sistema y la visualización de los datos recopilados por el servidor. Para acceder a esta interfaz, se ha implementado una autenticación automática mediante el uso de LDAP y el correo corporativo, consiguiendo la creación y mantenimiento automático de usuarios dentro del sistema sin necesidad de hacerlo manualmente. Finalmente, mediante el uso de Webhooks, se ha implementado un sistema automatizado para el envío de alertas cuando los datos monitorizados muestren valores anormales o críticos. En conclusión, el proyecto ha logrado implementar una solución completa y eficaz, Aquest treball s’ha realitzat amb l’empresa HP Inc., i sorgeix de la necessitat de redissenyar el sistema de monitorització d’una infraestructura de màquines en un entorn d’integració continua. L’objectiu es assolir una major automatització, portabilitat i escalabilitat del sistema, minimitzant així la intervenció manual i adaptant-se millor a majors volums de dades i dispositius. El projecte defineix el sistema actual i els problemes que presenta i, després d’un estudi de les diferents possibles solucions, detalla la solució proposada adaptant-se als requeriments del nou sistema. La solució proposada i posteriorment desenvolupada, es basa en la tecnologia Zabbix, i consta d'un servidor de monitoratge que emmagatzema i analitza les dades enviades pels hosts, les màquines que conformen el sistema i que es volen monitorar. Tant la creació del servidor, com la inserció dels hosts al sistema, es realitza mitjançant l'ús d'Ansible, tecnologia que permet l'automatització completa de la tasca. La recol·lecció de dades s'ha realitzat mitjançant la monitorització d'ítems de les màquines del sistema, com ara la càrrega de CPU o l'estat dels serveis que executen. Per això, s'han utilitzat ítems preestablerts per Zabbix i també s'ha desenvolupat un cicle de desenvolupament que permet crear ítems personalitzats. Zabbix ofereix una interfície gràfica per a la configuració del sistema i la visualització de les dades recopilades pel servidor. Per accedir a aquesta interfície, s'ha implementat una autenticació automàtica mitjançant l'ús de LDAP i el correu corporatiu, aconseguint la creació i el manteniment automàtic d'usuaris dins del sistema sense necessitat de fer-ho manualment. Finalment, mitjançant l'ús de Webhooks, s'ha implementat un sistema automatitzat per enviar alertes quan les dades monitorades mostrin valors anormals o crítics. En conclusió, el projecte ha aconseguit implementar una solució completa i eficaç per a la monitorització de la infraestructura de màquines en u, This project has been done with the enterprise HP Inc., and appears from the necessity to redesign the monitoring system of an IT infrastructure in a continuous integration environment. The objective is to improve the automation, portability and scalability of the system, thus minimizing manual intervention and making it adapt better to larger volumes of data and devices. The project defines the current system and the problems it presents and, after a study of the different possible solutions, details the proposed solution adapting to the requirements of the new system. The solution proposed and developed is based on Zabbix technology and consists of a monitoring server that stores and analyzes the data sent by the hosts, the machines that are part of the system and that needs to be monitored. Both the creation of the server and the insertion of the hosts into the system, are being done using Ansible, a technology that allows the automation of tasks. Data collection has been carried out by monitoring items on the system machines, such as the CPU load or the status of the services they run. For this, pre-established items by Zabbix have been used and a development cycle has also been developed that allows creating personalized items. Zabbix offers a graphical interface for the configuration of the system and to view the data collected by the server. To access this interface, automatic authentication has been implemented using LDAP and corporate email, achieving the automatic creation and maintenance of users within the system. Finally, through the use of Webhooks, an automated system has been implemented to send alerts when the monitored data shows abnormal or critical values. In conclusion, the project has managed to implement a complete and effective solution for monitoring an IT infrastructure in a continuous integration environment, achieving the established objectives.

This paper presents a new feature to enable Kokkos with transparent device selection. For application developers, it is not easy to identify which device is the most appropriate to use in a heterogeneous system, since this depends on the characteristics of both the application and the hardware. In Kokkos, a backend is associated with one specific programming model/hardware. Programmers decide which backend to use at compilation time. This new feature implemented on the OpenACC backend eliminates the burden of deciding which device to use, providing a highly productive programming solution for Kokkos applications. This work includes implementation details and a performance study conducted with a set of mini-benchmarks (i.e., AXPY and dot product), kernels (Lattice-Bolzmann method), and two mini-apps (LULESH and miniFE) on two heterogeneous systems with different hardware capabilities. This new Kokkos feature provides high accelerations of up to 35 × thanks to automatic and transparent device selection., This research used resources of the Oak Ridge Leadership Computing Facility and the Experimental Computing Laboratory at the Oak Ridge National Laboratory, which is supported by DOE’s Office of Science under Contract No. DE-AC05-00OR22725. This research was supported in part by the Exascale Computing Project (17-SC20-SC), a collaborative effort of the DOE’s Office of Science and the National Nuclear Security Administration. This material is based upon work by the RAPIDS Institute, supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, and Scientific Discovery through Advanced Computing (SciDAC) program. This manuscript has been authored by UT-Battelle LLC under Contract No. DE-AC05-00OR22725 with the DOE. The publisher, by accepting the article for publication, acknowledges that the US Government retains a non-exclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of the manuscript or allow others to do so, for US Government purposes. The DOE will provide public access to these results in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan)., Peer Reviewed, Postprint (author's final draft)

La plataforma de servicios Argos (https://www.hyds.es/ca/services/) es un sistema de alerta temprana y ayuda a la decisión, focalizado en la gestión de emergencias. Contiene múltiples fuentes de información: sensores, avisos oficiales, modelos de previsión, modelos específicos para fenómenos concretos, elementos vulnerables, llamadas 112, disrupciones del tráfico, etc. Las cuales se actualizan en tiempo real dando lugar a un flujo de gran volumen. Los usuarios de Argos son en general organismos involucrados en la gestión de las emergencias. En este proyecto se quiere desarrollar un módulo que permita simplificar la gestión multi-cuenta por parte de supervisores o comerciales. Este módulo tendrá que permitir la creación de nuevas cuentas, la modificación de las ya existentes, etc.. Se trabajará en colaboración con clientes para la definición y validación de los desarrollos realizados., The Argos service platform (https://www.hyds.es/ca/services/) is an early warning and decision support system focused on emergency management. early warning and decision support system focused on emergency management. It contains multiple sources of information: sensors, official warnings, forecasting models, specific models for specific specific models for specific phenomena, vulnerable elements, 112 calls, traffic disruptions, etc. These are updated in real time, resulting in a high-volume flow of information. high volume flow. Argos users are generally organizations involved in emergency management. emergency management. In this project we want to develop a module to simplify multi-account management by supervisors or sales representatives. This module will have to allow the creation of new accounts, the modification of existing ones, etc... We will work in collaboration with customers for the definition and validation of the developments made.

This thesis delves into the problem of compressing data generated by mobile networks, which are formed by many antennas and IoT devices, that emit huge quantities of measurements. Storing this data is growing more difficult and expensive, a trend driven by continuously escalating data transmission speeds, the sheer volume of traffic, and the increasing number of devices. To solve this problem, we surveyed some compression algorithms, both classical and learning-based, with a particular focus on deep-learning predictors applied in conjunction with entropy encoders. The core of our research is centered around compression algorithms based on Transformers and Arithmetic encoders. We propose several improvements like pretraining and using different models for each column, which outperform current approaches for our particular data domain. The focus of this research will be trying to improve compression on a large-scale dataset provided by Telefonica, which contains heterogeneous measurements from a huge network of antennas. Additionally, we provide a comprehensive analysis of the outcomes obtained, giving an intuition about why deep learning approaches work better than the current baselines, albeit at a higher computational cost, and explaining why our domain-specific solution slightly outperforms the rest.

Optical point-to-multipoint (P2MP) connectivity based on digital subcarrier multiplexing (DSCM) has been shown as a solution for the metro-access segment that is able to reduce capital and operational costs and support the capacity and high dynamicity needs of future 6G services. To achieve maximum performance, activation and deactivation of subcarriers must be done near-real-time to provide just the capacity needed to support the input traffic. In this paper, we investigate the applicability of various approaches capable of supporting the near-real-time operation requirement. Starting from the centralized approach that can be carried out on the centralized software-defined networking (SDN) controller, we also explore distributed approaches that might relieve the SDN controller from near-real-time operation. In particular, we explore the performance of deploying a multiagent system (MAS), where intelligent agents run on top of the nodes in the P2MP tree and communicate among them. Illustrative results show that the distributed approaches can achieve a performance close to that of the centralized one, while reducing communication needs. Results also show the importance of traffic/capacity prediction to anticipate the activation of subcarriers., The research leading to these results has received funding from the Smart Networks and Services Joint Undertaking under the European Union's Horizon Europe research and innovation programme under Grant Agreement No. 101096120 (SEASON), the MICINN IBON (PID2020-114135RB-I00) project and from the ICREA Institution., Peer Reviewed, Postprint (author's final draft)

En los últimos años, se han hecho descubrimientos en nanotecnología y materiales, que están dando paso a nanodispositivos con capacidades de detección, almacenamiento y procesamiento de datos. Se prevé que estos nanodispositivos naveguen por el torrente sanguíneo, con el fin de detectar eventos médicos de interés. La denominada "flow-guided localization" pretende asignar estos eventos a sus posiciones físicas, lo que permitiría un rápido diagnóstico y una reducción en costes e invasividad. Dado que este campo de investigación se encuentra aún en sus primeras fases, actualmente no hay un estándar común en las evaluaciones de este tipo de localización. Este hecho dificulta comparaciones objetivas entre este tipo de soluciones, lo que obstaculiza el avance en el ámbito. Como consecuencia, los investigadores se han centrado en el desarrollo de marcos estandarizados para evaluar los sistemas de "flow-guided localization". Sin embargo, aún no se ha investigado el impacto que la selección de los eventos tiene en su localización. Este estudio aborda esta laguna de conocimiento, analizando cómo la cantidad y la ubicación de los eventos a detectar afectan al rendimiento de los sistemas de localización. Para ello, se presenta una metodología de muestreo de puntos de evaluación, que garantiza estimaciones representativas de la precisión de estos algoritmos en todo el sistema cardiovascular. El conjunto de eventos obtenido reduce los recursos computacionales necesarios para las valoraciones, a la vez que mantiene la objetividad y fiabilidad de los resultados., Recent advances in the field of nanotechnology and novel materials are paving the way towards nanodevices that integrate sensing, storage and processing capabilities. These nanodevices are envisioned to navigate through the bloodstream, while sensing for medical events of interest. Flow-guided localization seeks to map these events to physical sites, resulting in early diagnosis while lowering costs and invasiveness. As the field is still in its early stages, existing evaluations of flow-guided localization are non-standardized, making objective comparisons challenging. To promote advancement in the field, researchers have focused on developing standardized frameworks for evaluating flow-guided localization systems. However, research into the impact of target event selection on localization is still lacking in the literature. This study addresses this knowledge gap by looking at how the number and placement of medical target events affect the performance of flow-guided localization systems. To address this, a methodology for sampling evaluation points is presented, ensuring representative assessments of performance indicators across the cardiovascular system. The derived subset of target events reduces the computational resources required for testing while maintaining objectivity and reliability of the results.

Donades les actuals condicions de crisi climàtica i, per tant, energètica, el control i optimització del consum de recursos naturals i tècnics ha esdevingut críticament necessari per a garantir la pervivència de les nostres societats. Alhora, aquest fet és especialment important pel que fa al sector HPC, que preveu un creixement sostingut i exponencial durant els anys vinents. És així que apareix la necessitat de dissenyar un programari que permeti optimitzar l'execució de les aplicacions que s'executen en aquests entorns, a fi de reduir-ne l'energia consumida. Aquest és el punt de partida d'Energy Aware Runtime, un software de sistema que permet monitorar i optimitzar l'execució d'aplicacions en entorns de HPC. El present treball estudiarà la millora de dos elements: per una banda, l'estratègia de classificació que, en temps d'execució, hauria de detectar la fase d'execució en què es troba l'aplicació en un moment donat, permetent així plantejar optimitzacions que siguin coherents. D'altra banda, es treballarà també en el criteri de diferenciació que, donades dues signatures d'una mateixa fase, permet saber si l'aplicació ha canviat suficient per a tornar a optimitzar-la. Aquest treball s'ha dividit en dues parts: per una banda, l'avaluació de la qualitat de les estratègies de classificació sobre un conjunt de dades caracteritzat prèviament i de forma orientativa, que s'ha articulat a través de la implementació d'una eina que automatitzés tot aquest procés, i, d'altra banda, la simulació del comportament dels diferents criteris de diferenciació considerats. Amb aquests dos processos experimentals, s'ha obtingut una classificació més robusta i coherent amb les característiques de l'arquitectura, així com un criteri que permetria optimitzar millor l'aplicació. Alhora, d'aquests dos resultats s'han plantejat les passes que s'haurien de seguir a fi d'implementar-les i utilitzar-les. Les principals conclusions d'aquest treball demostren la capacitat d'automatitzar l'ent, Given the current climate crisis and the subsequent energy crisis, the control and optimization of the consumption of natural and technical resources has become critically necessary to ensure the survival of our societies. Moreover, this has become particularly important for the HPC sector, which anticipates sustained and exponential growth in its energy consumption during the following years. Thus, it arises the need to design a software that allows to optimize the execution of applications running in these environments to reduce their energy consumption. This is the baseline of Energy Aware Runtime, a system software that allows the monitoring and optimizing of the execution of applications in HPC environments. The present work will study the improvement of two elements: on the one hand, the classification strategy which, at runtime, detects the execution phase in which the application finds itself at a given moment, thus allowing for coherent optimizations, and, on the other hand, the differentiation criterion which, given two signatures of the same execution phase, determines if the application has changed enough to be re-optimized. This work has been divided into two parts: on the one hand, the evaluation of the quality of the classification strategies over a previously characterized dataset, an experimental process that has been articulated and automated through the implementation of a tool, and, on the other hand, the simulation of the behavior of the different differentiation criteria considered. With these two experimental processes, a more robust and coherent classification has been obtained, as well as a new criterion that gives place to a better optimization of the application. At the same time, from these two results, a set of indications has been given in order to duly implement and use them. The main conclusions of this work demonstrate the ability to automate the training and incorporation of the different classification strategies, as well as the ca

In the digital era we are in, smartphones have become an indispensable part of our lives. They are not just communication devices but serve as gateways to a lot of important services: social media, commerce, banking, government apps, healthcare, etc. Given the importance of smartphones in the modern world, the security of mobile applications and mobile operating systems is really important. The focus of this project is to develop an obfuscator that aims to protect native applications that use the Java Virtual Machine, providing an additional layer of security against reverse engineering attempts.

Nuestro alumnado está en pleno proceso de maduración personal y mental en el que debemos ayudar, ya que es nuestro deber como profesorado y modelo. Para ello hay que incidir en ayudar a cada estudiante en la creación de su propia identidad como profesional de la informática. El primer paso consiste en aumentar su motivación por la informática como profesión y en aumentar su implicación en su propio desarrollo, empezando por el aprendizaje. Este artículo presenta unos principios para el diseño de una educación motivadora basados en lo que nos enseñan la literatura y ciertas experiencias de éxito., Our students are in the process of personal and mental maturation, and we must help them, as it is our duty as teachers and role models. In order to do this, we must focus on helping each student to create his or her own identity as a computing professional. The first step is to increase their motivation for computing as a profession and to increase their involvement in their own development, starting with learning. This article presents principles for the design of motivational education based on what the literature and some successful experiences teach us., Este trabajo ha contado con la financiación del Departament de Recerca i Universitats de la Generalitat de Catalunya – 2021 SGR 01412., Peer Reviewed, Objectius de Desenvolupament Sostenible::4 - Educació de Qualitat, Postprint (published version)

The rapid expansion of Machine Learning (ML) and Artificial Intelligence (AI) has profoundly influenced the technological landscape, reshaping various industries and applications. This surge in computational demands has led to the widespread adoption of Cloud data centers, crucial for supporting the storage and processing requirements of these advanced technologies. However, this expansion poses significant challenges, particularly in terms of energy consumption and associated carbon emissions. As the reliance on cloud data centers intensifies, there is a growing concern about the environmental impact, necessitating innovative solutions to enhance energy efficiency and reduce the ecological footprint of these computational infrastructures. This paper focuses on addressing the challenges linked to training ML and AI applications, emphasizing the importance of energy-efficient solutions. The proposed framework integrates components from the AI-SPRINT project toolchain, such as Krake, Space4AI-R, and PyCOMPSs. Our reference application involves training a Random Forest model for electrocardiogram classification, profiling available resources to obtain a performance model able to predict the training time, and dynamically migrating the workload to sites with cleaner energy sources providing guarantees on the training process due date. Results demonstrate the framework’s capacity to estimate execution time and resource requirements with low error, highlighting its potential for establishing an environmentally sustainable AI ecosystem., This work has been funded by the European Commission under the H2020 grant N. 101016577 AI-SPRINT: AI in Secure Privacy pReserving computINg conTinuum., Peer Reviewed, Postprint (author's final draft)

Recently, significant milestones have been achieved in the field of healthcare data analysis. However, alongside these accomplishments, substantial data-related challenges have emerged in the domain of big data management. Modern healthcare projects are no more dealing with a single data repository but many heterogeneous ones and must overcome data variety, privacy and governance issues. Yet, current solutions face a privacy-decentralization trade-off. To address this dual challenge, we introduce HealthMesh, a novel layered architectural framework based on the Data Mesh principles, providing a domaindecentralised paradigm. In addition, the framework incorporates a Semantic Data Model which establishes robust governance, enables interoperability and guarantees policy compliance for all the data assets. To demonstrate the capabilities of the proposed approach, we provide an illustrative example inspired by the use case of the INCISIVE project for breast cancer analytics. Overall, this work makes a significant contribution on collecting key challenges, identifying actors and providing a set of components and guidelines for establishing a holistic framework for the complex field of healthcare data management., The work reported here was supported by the EU-H2020 programme under GA.952179 INCISIVE, Horizon Europe Programme under GA.101095717 (SECURED) and GA.101135513 (CYCLOPS), the Spanish Research State Agency (MICINN/AEI, ERDF/FEDER) under GA.MCIN/AEI/10.13039/501100011033/FEDER (UE DALEST), the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00 / AEI/10.13039/501100011033 (DOGO4ML), and the Generalitat de Catalunya (AGAUR) 2021-SGR-00478 (CROMAI)., Peer Reviewed, Postprint (published version)

The main memory subsystem has traditionally been one of the more complex resources to analyze in multicore real-time embedded systems, with memory controller considerations and JEDEC timing constraints being the more prominent factors contributing to such complexity. One of the main challenges in multicore real-time systems is the production of the necessary evidence regarding the management of contention for the certification of multicore platforms in safety-relevant sectors. As current MPSoC platforms provide little information on how tasks may be interacting and delaying each other at large, it still remains a tall order to provide evidence about the correctness of hardware and software mechanisms deployed specifically to mitigate and manage contention on shared resources. This work attempts to bridge this gap by proposing a low-overhead hardware mechanism to tightly track inter-core contention within the main memory subsystem. The proposed technique enhances the quality of timing- and contention-related evidence, increasing the explainability and management of multicore contention in the main memory subsystem for multicore real-time systems in relation to applicable safety standards regulating their usage.

Earth system models (ESM) demand significant hardware resources and energy consumption to solve atmospheric chemistry processes. Recent studies have shown improved performance from running these models on GPU accelerators. Nonetheless, there is room for improvement in exploiting even more GPU resources. This study proposes an optimized distribution of the chemical solver's computational load on the GPU, named Block-cells. Additionally, we evaluate different configurations for distributing the computational load in an NVIDIA GPU. We use the linear solver from the Chemistry Across Multiple Phases (CAMP) framework as our test bed. An intermediate-complexity chemical mechanism under typical atmospheric conditions is used. Results demonstrate a 35× speedup compared to the single-CPU thread reference case. Even using the full resources of the node (40 physical cores) on the reference case, the Block-cells version outperforms them by 50%. The Block-cells approach shows promise in alleviating the computational burden of chemical solvers on GPU architectures., This work was partially supported by funding from the Ministerio de Ciencia, Innovación y Universidades as part of the BROWNING project (RTI2018-099894-BI00 funded by MCIN AEI/10.13039/ 501100011033) and part of the CAROL project (PID2020-113614RBC21 funded by MCIN AEI/10.13039/501100011033), the Generalitat de Catalunya (2021-SGR-00574, 2021-SGR-01550 and 2021- SGR00785) and the AXA Research Fund through the AXA Chair on Sand and Dust Storms established at the Barcelona Supercomputing Center (BSC). Matthew Dawson has received funding from the European Union’s Horizon 2020 research and innovation program under Marie Sklodowska Curie grant agreement no. 747048. Mario Acosta has received funding from the National Research Agency through OEMES (PID2020-116324RA-I00). This paper expresses the opinions of the authors and not necessarily those of the funding commissions., Peer Reviewed, Postprint (author's final draft)

Malware detection plays and important role in modern digital systems. Protecting against the fast-paced evolving cyber attacks is critical to safeguard sensitive information and preserve the integrity of digital infrastructure. Traditional signature-based detection methods are not effective when detecting new or altered versions of malware, such as polymorphic or metamorphic malware. Machine learning approaches have been proven to be much more effective at detecting such malware. Runtime behavior can be captured using the most fundamental part of a program, its instructions, also referred as the opcodes. This study presents both static and dynamic analysis using opcodes as the main feature for machine learning models., Funded by the European Union. Project number: 101093062., Preprint

In a world that is becoming increasingly globalized and, at the same time, more focused on a small group of companies, there is a need for software capable of carrying out the complex management required by the magnitude of these companies' operations. One of the most fundamental elements of the contemporary era is energy. This energy is generated in large facilities, and its management entails an even higher complexity than that of other companies, since it is not only necessary to be aware of the electrical generation of the company itself, but also of all those connected to the electrical grid. This work is part of the development of an infrastructure control software for the company Naturgy, which seeks to centralize the management and control of all its parks and power plants in a single place. To do this, multiple factors must be considered to design an error-proof system capable of not only controlling the current installations, but also of growing and scaling almost indefinitely. It is divided into two parts: on the one hand, a database and back-end part focused on the final product and, on the other hand, improvements in the quality of the development process, through automation, CI/CD, and documentation., En un món cada vegada més globalitzat i, alhora, més centrat en un petit grup d'empreses, neix la necessitat de comptar amb un programari capaç de realitzar les gestions complexes que la magnitud de les operacions d'aquestes empreses requereix. Un dels elements més fonamentals de l'era contemporània és l'energia. Aquesta energia és generada en grans instal·lacions, i la seva gestió entranya una complexitat encara més elevada que la d'altres empreses, ja que no només s'ha d'estar pendent de la generació elèctrica de la pròpia empresa, sinó de totes les connectades a la Xarxa Elèctrica. En aquest treball es troba part del desenvolupament d'un programari de control d'infraestructures per a l'empresa Naturgy, que busca centralitzar la gestió i control de tots els seus parcs i centrals en un únic lloc. Per fer-ho, cal tenir en compte múltiples factors i dissenyar un sistema a prova d'errors i capaç de, no només controlar les instal·lacions actuals, sinó també de créixer i escalar de manera gairebé indefinida. Està dividit en dues parts que figuren, d'una banda, una part de bases de dades i back-end enfocada al producte final i, d'altra banda, millores en la qualitat del procés de desenvolupament, de la mà d'automatismes, CI/CD i documentació.

We present an empirical approach to identify the key factors affecting the execution performance of task-based workflows on a High Performance Computing (HPC) infrastructure composed of heterogeneous CPU-GPU clusters. Our results reveal that the execution performance in distributed GPU-accelerated task-based workflows highly depends on several interrelated factors regarding the task algorithm, dataset, resources, and system employed. In addition, our analysis identifies key correlations among these factors, presents novel observations, and offers guidelines toward designing an automated method to handle task-based workflows in modern, high-compute capacity, CPU-GPU engines., This work has been partially supported by DEDS (H2020-MSCAITN2020) with grant agreement No. 955895, the EU-HORIZON programme CREXDATA under GA.101092749, the EU-HORIZON programme FAIR-CORE4EOSC under GA.101057264, the EUHORIZON programme EXTREMEXP under GA.101093164, the Spanish Government projects PID2019-107255GB and PID2020117191RB-I00/AEI/10.13039/501100011033andMCIN/AEI/10.13039 /501100011033 (CEX2021-001148-S), and by the Departament de Recerca i Universitats de la Generalitat de Catalunya (2021 SGR 00412, MPiEDist)., Peer Reviewed, Postprint (published version)

Checkpoint/Restore techniques had been thoroughly used by the High Performance Computing (HPC) community in the context of failure recovery. Given the current trend in HPC to use containerization to obtain fast, customized, portable, flexible, and reproducible deployments of their workloads, as well as efficient and reliable sharing and management of HPC Cloud infrastructures, there is a need to integrate Checkpoint/Restore with containerization in such a way that the freeze time of the application is minimal and live migrations are practicable. Whereas current Checkpoint/Restore tools (such as CRIU) support several options to accomplish this, most of them are rarely exploited in HPC Clouds and, consequently, their potential impact on the performance is barely known. Therefore, this paper explores the use of CRIU’s advanced features to implement diskless, iterative (pre-copy and post-copy) migrations of containers with external network namespaces and established TCP connections, so that memory-intensive and connection-persistent HPC applications can live-migrate. Our extensive experiments to characterize the performance impact of those features demonstrate that properly-configured live migrations incur low application downtime and memory/disk usage and are indeed feasible in containerized HPC Clouds., This research was partially supported by the Spanish Government under contract PID2019-107255GB-C22, by the Generalitat de Catalunya, Spain under contract 2021-SGR-00478, and by the EU-HORIZON programme under grant agreement 101092646., Peer Reviewed, Postprint (published version)

A common way of improving performance of applications for multi-core processors is to exploit parallelism. In deep learning (DL), training or tuning parameters use user’s sensitive data, and thus preserving privacy is critical. Hardware-assisted protection mechanisms (i.e., trusted execution environments - TEEs) offer a practical privacy-preserving solution, nowadays available both in private and public data centers. We present SGX-OmpSs, a new approach combining a task-based programming model (i.e., OmpSs) with TEEs (i.e., Intel Software Guard Extensions). SGX-OmpSs supports asynchronous task parallelism and hardware heterogeneity by using the data dependencies between tasks of the application, easily specified by code annotations. We evaluate SGX-OmpSs via several microbenchmarks and state-of-the-art DL applications and datasets (e.g., YOLO and MNIST). SGX-OmpSs achieves 94% gain speedup while offering additional security guarantees., Peer Reviewed, Postprint (author's final draft)

The main objective of this project is to develop a simulation environment for Unmanned Aerial Systems in Unity. This tool will improve the safety, operational efficiency, and training in the UAS sector. The application creates a simulation environment that can be operated by means of the following three different modes: Discover Mode, Simulation Mode, and Inverse Reality Mode. All of them use the Meta Quest 2 virtual reality headset to perform a more immersive experience.

© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Time-domain digital twin models for single carrier and DSCM signals are developed that propagate features to estimate the impact of filter penalties on the BER. Results show remarkable accuracy, which is used for lightpath provisioning., The research leading to these results has received funding from the European Community through the MSCA MENTOR (G.A. 956713), the European Union's Horizon Europe research and innovation programme SEASON (G.A. 101096120), the MICINN IBON (PID2020-114135RB-I00) and from the ICREA Institution., Peer Reviewed, Postprint (author's final draft)

Aquest Treball de Final de Grau, "Ampliació de funcionalitats a HuntDown: Integració d'un framework amb un sistema de decisions dinàmic" té com a objectiu aportar una millora a un software d'auditories de seguretat de programari lliure, HuntDown. El projecte es du a terme dins de la Universitat Politècnica de Catalunya (UPC) i és una continuació d'un projecte existent dirigit per Renè Serral Gracià. En concret, tracta la recerca i implementació d'un framework completament integrat a HuntDown i d'un motor de decisions bàsic, dotant l'eina d'una capacitat de decisió dinàmica. Aquest sistema serà crucial per facilitar a futurs desenvolupadors del projecte la substitució del motor per un de més complex, quan l'eina disposi de les funcionalitats pertinents., This Final Degree Project, entitled "Ampliació de funcionalitats a HuntDown: Integració d'un framework amb un sistema de decisions dinàmic" aims to enhance an open-source security audit system, HuntDown. The project is developed within the Universitat Politècnica de Catalunya (UPC) and serves as a continuation of an existing initiative led by Renè Serral Gracià. Specifically, it addresses the research and implementation of a fully integrated framework with the project, alongside a basic decision-making engine, endowing the tool with dynamic decision-making capabilities. This enhancement will be essential in aiding future developers of the project with the replacement of the engine for a more complex system, when the tool acquires the relevant functionalities.

Emerging from the symbiotic combination of nanotechnology and communications, the field of nanonetworking has come a long way since its inception more than fifteen years ago. Significant progress has been achieved in several key communication technologies as enablers of the paradigm, as well as in the multiple application areas that it opens. In this paper, the focus is placed on the electromagnetic nanonetworking paradigm, providing an overview of the advances made in wireless nanocommunication technology from microwave through terahertz to optical bands. The characteristics and potential of the compared technologies are then confronted with the requirements and challenges of the broad set of nanonetworking applications in the Internet of NanoThings (IoNT) and on-chip networks paradigms, including quantum computing applications for the first time. Finally, a selection of cross-cutting issues and possible directions for future work are given, aiming to guide researchers and practitioners towards the next generation of electromagnetic nanonetworks., S. A. acknowledges support from the EU’s Horizon Europe program through the European Research Council (ERC) under grant agreement 101042080 (WINC) and through the European Innovation Council (EIC) PATHFINDER scheme, grant agreement No 101099697 (QUADRATURE). C. H. acknowledges the in-part support from the National Natural Science Foundation of China (NSFC) – European Research Council (ERC) Research Program under Project No. 62311530342, and from Alexander von Humboldt Foundation. The EU partially supported the activity of L. G. under the Italian National Recovery and Resilience Plan (NRRP) of NextGenerationEU, partnership on “Telecommunications of the Future” (PE0000001 - program “RESTART”). I. F. A. is supported by Rannis (Icelandic Research Fund) for the Grant of Excellence program at the University of Iceland. J. M. J. acknowledges the support of the US National Science Foundation through Award CNS-1955004, Award CNS-2011411, Award CNS-2225590, and CBET-2039189., Peer Reviewed, Postprint (author's final draft)

Aquesta tesi presenta EAR Data Analyzer, una nova eina dissenyada per a extreure, processar i analitzar dades obtingudes durant la "Learning Phase" en el marc de Energy Aware Runtime (EAR). La "Learning Phase" és un període crític per a la recopilació de dades relacionades amb el consum d'energia i el comportament del sistema. No obstant això, aquesta fase sovint s'enfronta a desafiaments a causa de la presència de dades inconsistents o errònies, que poden impedir una anàlisi precisa. EAR Data Analyzer aborda aquest problema proporcionant una solució integral per a identificar anomalies en el conjunt de dades recopilades. L'objectiu principal d'aquest treball és el desenvolupament i la implementació d'EAR Data Analyzer, posant l'accent en la seva funcionalitat per a detectar irregularitats, inclosos valors incoherents i punts de dades errònies. L'eina facilita la identificació de valors atípics i discrepàncies, la qual cosa permet als administradors d'EAR a refinar les dades de manera eficaç. A més, veurem la creació d'un Grafana Dashboard dedicat, mostrant la importància de proporcionar una interfície fàcil d'usar per a interpretar les dades generades. En resum, EAR Data Analyzer es perfila com una valuosa addició en l'àmbit de la informàtica amb consciència energètica, oferint un enfocament sistemàtic per a validar, refinar i visualitzar les dades obtingudes durant la "Learning Phase". La seva aplicació en el marc d'EAR subratlla el seu potencial per a millorar significativament la precisió de les anàlisis de consum d'energia, aplanant el camí per a una presa de decisions més informades i una gestió eficient del sistema., This thesis introduces EAR Data Analyzer, a novel tool designed to extract, process, and analyze data obtained during the Learning Phase within the framework of Energy Aware Runtime (EAR). The Learning Phase serves as a critical period for gathering data related to energy consumption and system behaviors. However, this phase often confronts challenges due to the presence of inconsistent or erroneous data, which can impede accurate analysis. EAR Data Analyzer addresses this issue by providing a comprehensive solution for identifying anomalies within the gathered dataset. The primary focus of this work involves the development and implementation of EAR Data Analyzer, emphasizing its functionality in detecting irregularities, including inconsistent values and erroneous data points. The tool facilitates the identification of outliers and discrepancies, enabling EAR administrators to refine the data effectively. Furthermore, we will go through the creation of a dedicated Grafana dashboard, showcasing its significance in providing a user-friendly interface for interpreting the data generated. In summary, EAR Data Analyzer emerges as a valuable addition to the realm of energy-aware computing, offering a systematic approach to validate, refine, and visualize data obtained during the Learning Phase. Its implementation within the EAR framework underscores its potential to significantly improve the accuracy of energy consumption analyses, paving the way for more informed decision-making and efficient system management.

Els certificats digitals al protocol TLS juguen un paper crucial per mantenir la confidencialitat i seguretat de les comunicacions digitals. Tot i la seva importància, l'emissió errònia o maliciosa de certificats digitals representa una vulnerabilitat significativa. Els atacs de phishing i l'emissió fraudulenta de certificats poden comprometre la confiança de l'usuari i la seguretat de les comunicacions en línia. Aquests problemes sorgeixen de les limitacions en la manera com les Autoritats de Certificació (CA) emeten i validen certificats, cosa que pot resultar en certificats erronis o maliciosos. Per abordar aquestes vulnerabilitats, es va desenvolupar el concepte de Registres de Transparència de Certificats (CT Logs). Aquests registres públics i accessibles contenen informació sobre tots els certificats emesos, permetent una auditoria oberta i la detecció primerenca de certificats sospitosos o maliciosos. Aquest projecte proposa un servei de monitorització dels Registres de Transparència de Certificats anomenat CT Watcher que ofereix als seus clients visibilitat dels certificats emesos i seguretat davant de possibles atacs., Digital certificates in the TLS protocol play a crucial role in maintaining the confidentiality and security of digital communications. Despite their importance, the erroneous or malicious issuance of digital certificates represents a significant vulnerability. Phishing attacks and fraudulent issuance of certificates can compromise user trust and the security of online communications. These problems arise from limitations in the way Certificate Authorities (CA's) issue and validate certificates, which can result in erroneous or malicious certificates. To address these vulnerabilities, the concept of Certificate Transparency Logs (CT Logs) was developed. These public and accessible logs contain information on all issued certificates, enabling open auditing and early detection of suspicious or malicious certificates. This project proposes a Certificate Transparency Logs monitoring service called CT Watcher that offers its customers visibility of issued certificates and security against possible attacks.

Aquest projecte es centra en desenvolupar i implementar un mètode d'autenticació innovador basat en blockchain per a les comunicacions IoT (Internet of Things). L'objectiu és millorar la seguretat dins de les xarxes IoT mitjançant l'ús de la tecnologia blockchain per a una autenticació de dispositius robusta i fiable. La implementació mostra un enfocament pràctic per abordar els desafiaments de seguretat en els entorns IoT, oferint una solució escalable i eficient per garantir la integritat de les dades i la confiança en les comunicacions IoT., This project is delivered as the Final Project of Computer Engineering Degree of the Barcelona School of Informatics. It is intended to put into practice the knowledge acquired during the mention in Information Technology. This project is focused on developing and implementing a novel blockchain-based authentication method for IoT (Internet of Things) communications. The objective is to enhance security within IoT networks by leveraging blockchain technology for robust and reliable device authentication. The implementation showcases a practical approach to addressing the security challenges in IoT environments, offering a scalable and efficient solution for ensuring data integrity and trust in IoT communications.

En aquest projecte s'ha buscat utilitzar una de les eines més potent d'avui en dia com és Blockchain i aprofitar les seves propietats per donar servei el món de la Intel·ligència Artificial. La idea és fer servir les propietats d'emmagatzematge que proporciona Blockchain per a crear un entorn de compartició de models d'Intel·ligència Artificial entre els diferents nodes participants. Aquests nodes han de poder interactuar amb el DLT (Distributed Ledger Technology) per a guardar els seus models entrenats i descarregar els que altres nodes han guardat. En aquest treball s'inclou l'anàlisi de la viabilitat del projecte, el disseny, l'experimentació per a l'obtenció de resultats i conclusions, i finalment la implementació la solució òptima. Malgrat que m'he trobat amb limitacions i dificultats tècniques, he aconseguit arribar a una solució viable i amb bons resultats, In this project, the aim has been to use one of today's most powerful tools, which is Blockchain, and leverage its properties to provide a service to the world of Artificial Intelligence. The idea is to use the storage properties provided by Blockchain to create an environment for sharing Artificial Intelligence models among different participating nodes. These nodes can interact with the DLT (Distributed Ledger Technology) to store their trained models and download those that other nodes have saved. This includes analyzing the viability of the project, designing, experimenting to obtain results and conclusions, and finally implementing the optimal solution. Despite encountering technical limitations and difficulties, I have arrived at a viable solution with good results.

The AI-SPRINT project, launched in 2021 and funded by the European Commission, focuses on the development and implementation of AI applications across the computing continuum. This continuum ensures the coherent integration of computational resources and services from centralized data centers to edge devices, facilitating efficient and adaptive computation and application delivery. AI-SPRINT has achieved significant scientific advances, including streamlined processes, improved efficiency, and the ability to operate in real time, as evidenced by three practical use cases. This paper provides an in-depth examination of these applications – Personalized Healthcare, Maintenance and Inspection, and Farming 4.0 – highlighting their practical implementation and the objectives achieved with the integration of AI-SPRINT technologies. We analyze how the proposed toolchain effectively addresses a range of challenges and refines processes, discussing its relevance and impact in multiple domains . After a comprehensive overview of the main AI-SPRINT tools used in these scenarios, the paper summarizes of the findings and key lessons learned., Peer Reviewed, Postprint (published version)

Aquest Treball de Final de Grau se centra en la migració de l'Argos, un visor web multi-hazard proporcionat per hydrometeorological innovative solutions (hyds) utilitzat per diverses agències de gestió d'emergències. En un context en què l'empresa ha prioritzat el processament i visualització de les dades, aquest projecte es concentra en modernitzar l'arquitectura de l'aplicació, migrant aquesta a un entorn modular, utilitzant Vue.js com a framework pel frontend, i .NET per desenvolupar una API que actuï d'intermediari entre la base de dades i el frontend. La finalització d'aquest projecte dotarà a l'empresa d'experiència valuosa per futures migracions d'altres serveis basats en la mateixa arquitectura monolítica., This Final Project focuses on the migration of Argos, a multi-hazard web viewer provided by hydrometeorological innovative solutions (hyds) and used by several emergency management agencies. In a context in which the company has prioritised data processing and visualisation, this project focuses on modernising the architecture of the application, migrating it to a modular environment, using Vue.js as the frontend framework, and .NET to develop an API to act as an intermediary between the database and the frontend. The completion of this project will provide the company with valuable experience for future migrations of other services based on the same monolithic architecture.

En un món on les tecnologies avancen ràpidament, la ciberseguretat esdevé un aspecte cada vegada més crític en aquesta època i en les futures. D'altra banda, amb la creixent digitalització dels processos i la informació a les empreses, l'eficiència dels ERP es torna fonamental per gestionar aquests recursos de manera efectiva. Aquest projecte final de grau reconeix la importància de la seguretat en un entorn empresarial que està en constant evolució. En el procés d'estudiar el context de seguretat, es busca reforçar la protecció de la informació gestionada per l'ERP mitjançant la identificació de vulnerabilitats i la proposició de solucions robustes. En aquest cas, l'enfocament se centra a Odoo, popular en l'àmbit de la gestió empresarial, amb l'objectiu d'assegurar la integritat i confidencialitat de les dades, així com d'agilitzar els processos relacionats amb la seguretat del sistema. Es busca contribuir a la identificació, creació i implementació de processos i bones pràctiques per millorar la seguretat dels sistemes ERP. L'objectiu es basa a oferir una anàlisi exhaustiva i tècnica, mitjançant propostes concretes per adaptar-se als reptes actuals i futurs de la ciberseguretat empresarial., In an environment where technologies advance so quickly, cyber security becomes an increasingly critical aspect in this era and in the future. Instead, with the increasing digitization of processes and information in companies, the efficiency of ERPs becomes fundamental to manage these resources effectively. This final degree project recognizes the importance of security in a business environment that is constantly evolving. In the process of studying the security context, the objective is to strengthen the protection of the information managed by the ERP through identification of vulnerabilities and the proposal of robust solutions. In this project, the focus is on Odoo, enterprise resource planning software used company-wide for the management of business processes, with the aim of ensuring data integrity and confidentiality on this software, as well as streamlining processes related to system security . With this work, I intent to identify, create and implement best practices and processes to improve security in ERP systems. The purpose is to offer a comprehensive and technical analysis through concrete proposals to adapt to current and future business cyber security challenges.

This work is focused on the transparent execution of Cellular Automata models on a multi-GPU architecture. Although Cellular Automata models can be easily parallelized on a single GPU, the domain size and transition function complexity may require the use of multiple GPUs. Our goal is to allow modellers to be completely unaware of the parallel execution context, i.e., the code implementing the Cellular Automata model remains the same regardless if the execution is performed on CPU, single GPU, or multi-GPU systems. This paper supplies meaningful technical insights on how to ensure both transparency and efficiency in multi-GPU execution of Cellular Automata models. In particular, an object-oriented approach is exploited in which a transparent layer is devised that abstracts the parallelization details and allows a strong “separation of concerns” between the execution parallelism issues and the model implementation. Preliminary experiments have been carried out on the multi-GPU cluster CTE-POWER available at the Barcelona Supercomputing Center (BSC), witnessing good speedups notwithstanding the transparency feature supplied by our approach., This research was partially funded by the Italian “ICSC National Center for HPC, Big Data and Quantum Computing” Project, CN00000013 (approved under the Call M42C – Investment 1.4 – Avviso “Centri Nazionali” - D.D. n. 3138 of 16.12.2021, admitted to financing with MUR Decree n. 1031 of 06.17.2022). This work was funded by the Next Generation EU - Italian NRRP, Mission 4, Component 2, Investment 1.5, call for the creation and strengthening of ’Innovation Ecosystems’, building ’Territorial R&D Leaders’ (Directorial Decree n. 2021/3277) - project Tech4You - Technologies for climate change adaptation and quality of life improvement, n. ECS0000009. This work was supported by Italian Ministry of Research (MUR) under PNRR projects FAIR “Future AI ResearcH” - Spoke 9 - CUP H23C22000860006. This research was partially funded by the Spanish Ministry of Education (PID2019-107255GB-C22) and the Generalitat de Catalunya (2021-SGR-01007)., Peer Reviewed, Postprint (author's final draft)

El treball presenta el desenvolupament d'una plataforma que té com a objectiu ajudar amb el desafiament de la despoblació rural a través de la connexió de talents locals amb oportunitats professionals a les seves regions d'origen. Amb el propòsit de disminuir la tendència migratòria cap a àrees urbanes, la plataforma possibilita la interacció entre estudiants, professionals i la població general que han abandonat les zones rurals, i les empreses, associacions i administracions públiques que busquen talent local amb desig de tornar. La present eina no sols persegueix fomentar el retorn dels talents als seus llocs d'origen, sinó també es posiciona com un catalitzador per a la diversificació econòmica i la innovació en les àrees rurals, alineant-se amb les Agendes Internacionals de Desenvolupament Sostenible i els objectius i projectes de diversos actors de les diverses àrees rurals de l'Estat. La implementació de la plataforma requereix la creació i el desenvolupament d'aplicacions web i mòbils, així com la creació d'una infraestructura robusta i escalable que possibilita un increment potencial de la demanda d'aquesta, així com l'ús simultani de múltiples usuaris. Aquesta iniciativa no sols tracta d'una qüestió crítica per al desenvolupament sostenible de les àrees rurals, sinó que també ofereix una oportunitat per a aplicar els coneixements adquirits durant el màster, combinant habilitats tècniques amb una comprensió exhaustiva de les demandes socials i econòmiques de l'entorn. Com a part del projecte, es recalca la importància de la tecnologia com a eina per a la cohesió territorial i el desenvolupament local, oferint una solució nova que pot ser replicada en altres contextos similars.

This final thesis project delves into the persistent challenge of bug prevention in software engineering. Throughout numerous existing techniques such as dynamic analysis with fuzzers, our study narrows its focus on static code analyzers - tools that approach the analysis on the actual source code, without the need to execute the program. Building upon prior research that underscored the effectiveness of dynamic analysis with fuzzers in the context of Linux kernel exploitation, this study investigates the advantages of static code analyzers with a clear goal; to design and develop a new static code analyzer tool (Apokalypto) to unveil bugs hitherto unknown.

In this master's thesis, three systems for an unmanned surface vehicle (USV) are designed, developed, and tested. These systems include a collision avoidance system, a method to create 3D models of underwater structures, and an integration system for a USV and a remotely operated vehicle (ROV). The first system utilizes LiDAR technology along with four RGB cameras to detect the location of objects around the USV. Based on this information, received via an MQTT channel, the USV defines protection zones for each target. If the USV's protection zone intersects with that of a detected object, it enters a collision alert state and calculates an evasive manoeuvre in accordance with COLREGs. The COLREGs (Convention on the International Regulations for Preventing Collisions at Sea) are international rules established by the International Maritime Organization (IMO) to prevent collisions at sea. This system enhances the autonomy of the USV by enabling it to avoid obstacles in its path. The second system facilitates the scanning of submerged structures using an ROV and then reconstructing them into 3D models. Three methods were investigated for this purpose: stereoscopy, structure from motion (SFM), and neural radiance fields (NeRF). Testing various programs and cameras revealed that both SFM and stereoscopy are effective options for creating 3D models of underwater structures. The third system involves the transportation and remote control of an ROV using a USV, leveraging the USV's onboard computer. This system includes a winch and a framework that secures and tows the ROV safely. The entire system is operated remotely using the USV's remote control. These three systems have been developed for a smaller USV, with the goal of future implementation in a larger USV to improve performance. This approach allows the testing and refinement of these systems on a currently robust USV. This work covers a wide range of topics, including data processing, MQTT communication, image proce, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura, Objectius de Desenvolupament Sostenible::14 - Vida Submarina

The Dagstuhl Seminar 23352, titled “Integrating HPC, AI, and Workflows for Scientific Data Analysis,” held from August 27 to September 1, 2023, was a significant event focusing on the synergy between HighPerformance Computing (HPC), Artificial Intelligence (AI), and scientific workflow technologies. The seminar recognized that modern Big Data analysis in science rests on three pillars: workflow technologies for reproducibility and steering, AI and Machine Learning (ML) for versatile analysis, and HPC for handling large data sets. These elements, while crucial, have traditionally been researched separately, leading to gaps in their integration. The seminar aimed to bridge these gaps, acknowledging the challenges and opportunities at the intersection of these technologies. The event highlighted the complex interplay between HPC, workflows, and ML, noting how ML has increasingly been integrated into scientific workflows, thereby enhancing resource demands and bringing new requirements to HPC architectures, like support for GPUs and iterative computations. The seminar also addressed the challenges in adapting HPC for large-scale ML tasks, including in areas like deep learning, and the need for workflow systems to evolve to leverage ML in data analysis fully. Moreover, the seminar explored how ML could optimize scientific workflow systems and HPC operations, such as through improved scheduling and fault tolerance. A key focus was on identifying prestigious use cases of ML in HPC and understanding their unique, unmet requirements. The stochastic nature of ML and its impact on the reproducibility of data analysis on HPC systems was also a topic of discussion., Postprint (published version)

Despite the large impact chronic obstructive pulmonary disease (COPD) that has on the population, the implementation of new technologies for diagnosis and treatment remains limited. Current practices in ambulatory oxygen therapy used in COPD rely on fixed doses overlooking the diverse activities which patients engage in. To address this challenge, we propose a software architecture aimed at delivering patient-personalized edge-based artificial intelligence (AI)-assisted models that are built upon data collected from patients’ previous experiences along with an evaluation function. The main objectives reside in proactively administering precise oxygen dosages in real time to the patient (the edge), leveraging individual patient data, previous experiences, and actual activity levels, thereby representing a substantial advancement over conventional oxygen dosing. Through a pilot test using vital sign data from a cohort of five patients, the limitations of a one-size-fits-all approach are demonstrated, thus highlighting the need for personalized treatment strategies. This study underscores the importance of adopting advanced technological approaches for ambulatory oxygen therapy., This research was funded by the Spanish Ministry of Science, Innovation and Universities and FEDER, grant number PID2021-124463OB-100, and by the AGAUR Catalan Agency, grant number 2021_SGR_00326., Peer Reviewed, Postprint (published version)