PRR Project

Summary

Data are now considered an economic resource as valuable as oil. This has an increasing impact in science and technology, and even in people´s daily lives. LLMs (Large Language Models) are now part of the public imagination, along with ChatGPT. Less well known to most people, Digital Twins are seen as the next big innovation in science and industry, for instance, to enable the transition to renewable energy sources and to more sustainable consumption patterns.The dominant paradigm underlying the Big Data revolution is that of cloud computing services, where it is used to support an ecosystem of related applications. In this approach, known as Web 2.0, each organization constitutes a closed silo, leading to control concerns and limiting opportunities for innovation. This privileged access represents a significant competitive advantage for installed players, as it is precisely the ability to develop and operate infrastructures at scale that also translates into leadership for large companies in this area. As a reaction to the loss of control over data and encouraged by the success of cryptocurrencies, the proposal for radical decentralization of data management and applications known as Web 3.0 has been advocated. However, this perspective is not realistic in the short or medium term in terms of efficiency and scale. The alternative proposal is that of a single data market in Europe, centered on the concept of data spaces. In this case, there is decentralization of data production and consumption to foster both sovereignty over data and economic competition and technological innovation. Unlike Web 3.0, which is based solely on a technological solution, here there is the robust support of an ecosystem of laws, regulations, and standards emerging from a democratic and collaborative process. The technological support for data storage, transfer, and sharing currently proposed is still very limited compared to the systems developed for Web 2.0.Research in systems aspects of data management thus stems from the observation that (i) the first challenge in exploiting large amounts of data is the very scale and technical complexity of the infrastructures; and that (ii) current decentralized proposals are still far from being able to compete with Web 2.0. This is worrying because progress in data exploitation is limited, above all, by the availability of data and the corresponding infrastructures (https://www.spacemachine.net/views/2016/3/datasets-over-algorithms). The research question that justifies the proposed job is, thus: How do data management systems have to evolve towards a sustainable, efficient, and resilient single data market and even, in the long term, towards more radical decentralization? This is therefore a project that focuses on non-functional aspects of data management: fault tolerance, scale, performance, and security. These issues are addressed mainly with operating systems and distributed systems techniques. Our research on development in this area is directly contributing to Sustainable Development Goals of the United Nations 2030 Agenda 7, 9, and 12 in the context of projects such as: GreenDat.ai - Energy-efficient AI-ready Data Spaces (Horizon 101070416); CircThread - Building the Digital Thread for Circular Economy Product, Resource & Service Management (Horizon 958448); TwinEU - Digital Twin for Europe (Horizon 101136119); and Sustainable HPC - Highly Sustainable Performance Computing (FAI/FEE). DI and HASLab have hosted, in the scope of these projects, a number of young researchers / teaching assistants that are key to their success and would be eligible for an FCT Tenure position.In this context, candidates should match the following profile, being able to:- make original contributions to the state of the art in distributed systems and operating systems, in particular, with applications to data management; - lead research teams, in particular, in the context of collaborative research projects and the application project funding entities; - teach and supervise students in topics related to distributed systems and operating systems; - participate in development and technology transfer activities towards achieving wider economic and societal impact.It is expected that the prospective employee is assigned the following tasks:- development of topics related to non-functional aspects of data management in the teaching curricula of operating systems and distributed systems courses at bachelor and master levels; - supervision of students and junior researchers on topics related to non-functional aspects of data management; - contribution to research on non-functional aspects of data management within the scope of operating systems and distributed systems fundamentals, techniques, and tools; - promotion of novel research and development opportunities that advance the state of the art and realize its impact in industry and society.