Current PhD Students

Luís Ferreira da Silva, M.Sc. (FCT/UNL)Supervisor: Fernando Brito e AbreuProgram: DIN (FCT/UNL)Thesis title: A Pattern-Based Approach to Scaffold the IT Infrastructure Design ProcessAbstract: The design of Information Technology (IT) infrastructures is a challenge task since it implies a depth of knowledge in several areas that rarely reside in a single person. Most IT infrastructure designs, are based on proprietary models known as “blueprints” or product-oriented architectures defined by vendors to facilitate the configuration of a particular solution, based upon their services and products portfolio. Although existing “blueprints” can be facilitators in the solutions’ design, most of them do not allow the use of technologies from different vendors, making the decision process more complex and limiting infrastructure design to product features, provided by a given product or a family of products offered by a specific vendor. Additionally, these “blueprints” have a short lifecycle since they are usually associated with a particular version of a product or technology, which makes the possibility of using "blueprints" as a tool for the re-use of knowledge for recurring very limited. The main research problems are, (i) the inability to reuse knowledge in terms of best practices in the design of IT infrastructures, (ii) how to prove that the existent IT infrastructure knowledge is in fact relevant that are worth to be reused and, (iii) how to simplify the usage of this knowledge, making the IT infrastructure design simpler, quickly, better documented, facilitating the integration of components from different vendors and minimizing the communication problems between teams. The research goals consists in the (i) formalization of a vendor-independent pattern language able to support the process of designing IT Infrastructures, (ii) detection and recognition of implemented IT infrastructure patterns in order to prove that this knowledge exists and can be reused (iii) development of a computational environment that simplifies the use of design patterns to support the selection of the best patterns to solve a specific problem.

Sérgio Bryton, M.Sc. (FCT/UNL)Supervisor: Fernando Brito e AbreuProgram: DIN (FCT/UNL)Thesis title: Refactoring Process Improvement: Opportunities, Alternatives and BenefitsAbstract: Refactoring, in spite of widely acknowledged as a best practice, still has some limitations; namely, it lacks evidence regarding its claimed benefits on software quality, and tasks like the identification of refactoring opportunities and the decision on how to refactor are still performed manually and arbitrarilly, making this a cumbersome, irrepeatable and error-prone process and, ultimately, hampering its widespread adoption by the community as a regular practice. This research will address these limitations by providing, on one hand, quantitative evidence on the effects of refactoring on software quality and, on the other hand, providing a method and techniques which are expected to support tool-based refactoring activities. By achieving these results, we expect to clarify the relevance of refactoring to software quality and reduce the time and cost actually put into this activity, thus facilitating its widespread adoption.

Pedro Monteiro, M.Sc. (FCT/UNL)Supervisor: Miguel Pessoa MonteiroProgram: DIN (FCT/UNL)Thesis title: HARPPIE: Hyper Algorithmic Recipe for Productive Parallelism Intensive EndeavorsAbstract: Over the last few years, Parallelism has been gaining increasing importance and multicore processing is now common. Massification of parallelism is driving research and development of novel techniques to overcome current limits of Parallel computing. However, the scope of parallelization research focuses mainly on ever-increasing performance and much still remains to be accomplished regarding improving productivity in the development of parallel software. This PhD research aims to develop methods and tools to dilute parallel programming complexity and enable non-expert programmer to fully benefit from a new generation of parallelism-driven programming platforms. Although much work remains to be done to reduce the skill requirements for parallel programming to become within reach of medium-skill programming workforces, it is our belief that this research will help bridge that gap.

Carlos Fábio Ramos Conceição, M.Sc. (UNIFACS)Supervisors: Glauco Carneiro / Fernando Brito e AbreuProgram: DCTI (ISCTE-IUL)Thesis title: Collaborative Environments Based on Multiple Interactive ViewsAbstract: My research is about the use of collaborative elements to enrich Software Engineering environments based on Multiple Interactive Views. The collaborative elements cover coordination, communication, cooperation and awareness aspects and are expected to improve the support for software comprehension activities undertaken collectively by groups operating in a distributed manner. Empirical evidences will be sought to corroborate the expected improvements.