Optimizing the performance of micro-service and FaaS implementations for new hardware architectures. Given that the implementation of micro-services and FaaS backends will increasingly run on new hardware architectures, namely those based on a PM memory tier that can be leveraged for performance of these micro-services or functions, then optimizing kernel-level page migration for persistent memory becomes an important goal. We will study possible improvements to page migration in Linux when performed with the move_pages function (from libnuma), including: a) Exploiting parallel data copy, using either parallel threads (in multicores) and/or parallel DMA channels. b) Tuning the above method according to the asymmetric bandwidth bottlenecks of each memory type (e.g., more threads/DMA channels when migrating up vs. fewer when migrating down). c) Studying other overheads in move pages, e.g. related to page table locking and cache efficiency. The tasks above require experimentally studying and incorporating components inside Linux kernel (kernel programming). The expected outcome is a design and implementation of an improved move pages alternative, together with a comprehensive evaluation comparing it to the standard implementation of move pages, including benchmarks from typical micro-service and FaaS workloads.

The student will join the project DIVINA, Detecting Injection Vulnerabilities in Node. Applications, in the scope of its second task, whose main goal is to build a new platform for detecting and confirming injection vulnerabilities in Node. The student will develop an automatic symbolic test generator for confirming potential injection vulnerabilities detected by a separate static analysis, currently being developed by other project members. The symbolic tests to be generated will be executed in the JaVerT tool, which will have to be extended with new symbolic reasoning facilities.

The accepted candidates will work on improving the interoperability between C++ and Rust through static analysis techniques. There will be a strong focus on scalability, with the goal of scaling to very large codebases.

The work plan consists on: studying the problem and related work on blood collection; studying pattern mining approaches for temporal pattern discovery and spatio-temporal pattern discovery; analysing exploratory data using IPST blood collection data; proposing a machine learning methodology to unravel temporal and geographic patterns from IPST blood collection data; implement the pattern discovery proposed methodology; using the pattern discovery methodology implemented to uncover temporal and geographic patterns from IPST blood collection data; and evaluating, ranking and interpreting the results.

The goal of this scholarship is to develop a system to evaluate the Goodness of Prosody of non-native speakers. This will include using classifiers for different types of prosodic events, such as boundaries, stresses, and tones. These shall be tuned using different types of data: first only native data of the same variety, then native data of different varieties, then non-native but proficient data. Finally, the system will be assessed by comparing its results to annotated non-native data.

Development of deep learning based diagnostics from medical images, with a focus on digital pathology and an emphasis on breast cancer. The successful applicant will be able to apply his/her knowledge of deep learning to the problem of diagnosing breast cancer from biopsies

The main objective of this contract is to carry out research on efficient algorithms, architectures and tools specifically designed for maximising the performance and energy efficiency of emerging HPC systems. Special emphasis is given to sparse matrix, tensor and graph computations, while also opening up new application areas in data analytics and deep learning. This position is open in the scope of the European Project Sparcity (http://sparcity.eu/ - Euro-HPC project No 956213). Opportunities for performing high quality research, as part of a well known and highly motivated consortium of international researchers and students. Possibilities to collaborate with international industry and academic partners, and perform research visits to other universities or labs. For more information go to: https://www.euraxess.pt/jobs/797038

International selection call for one doctorate position in the scope of the project UIDP/50021/2020, funded by Fundação para a Ciência e a Tecnologia, in the form of an employment contract under an uncertain term work contract – in the framework of abovementioned legislation being the basis of the contracting the performance of a specific service, precisely defined and non-durable , with a view to performing the following functions: a) Contribute to the technological development of national and international research projects b) Manage the activities to be developed under the projects being executed c) Assure the respect of the deadlines within the projects execution d) Organise and coordinate meetings of the projects (internal meetings, consortium meetings, meetings with the European Commission, etc) e) Collaborate in the definition of communication and dissemination activities as well as exploitation activities f) Collaborate in the development of new proposals For more information go to: https://www.euraxess.pt/jobs/819742

The main objective of this contract is to carry out research on efficient algorithms, architectures and tools specifically designed for maximising the performance and energy efficiency of emerging HPC systems. Special emphasis is given to sparse matrix, tensor and graph computations, while also opening up new application areas in data analytics and deep learning. This position is open in the scope of the European Project Sparcity (http://sparcity.eu/ - Euro-HPC project No 956213). Opportunities for performing high quality research, as part of a well-known and highly motivated consortium of international researchers and students. Possibilities to collaborate with international industry and academic partners, and perform research visits to other universities or labs. For more information go to: https://www.euraxess.pt/jobs/797036

Te@ch4SocialGood aims to develop, pilot test, and study the efficacity of a teacher training program to promote both engagement as well as competence on the subject of cyberbullying. The grant holder will work on an embodied conversational agent integrated in a multi-modal learning platform supporting the teachers’ learning process. This will entail research and implementation of state-of-the art techniques in intelligent virtual agents.

Te@ch4SocialGood aims to develop, pilot test, and study the efficacity of a teacher training program to promote both engagement as well as competence on the subject of cyberbullying. The grant holder will work on the serious game and the content management platform supporting the learning process. This includes working on the serious game being developed, creating a learner profile by capturing data while the game is being played and activities are performed on the platform, using the learner profile to suggest new content to the learner.

The main goal is to develop new (e.g. linear programming) models and algorithms to optimize blood collection planning to better match transfusion needs in mainland Portugal, in order to deliver the right blood products when they are needed, avoiding dramatic consequences of shortages and ethical implications of waste.

- Develop a simulation platform based in multi-agents’ systems allowing the simulation of the impact of mass development of electric vehicles in smart cities. - Test electric vehicles management strategies.

-Develop tools allowing the optimal management of V2X considering the integration in different type of buildings. - Test electric vehicles management strategies.