Proposta de dissertação do MEI
Título: Data-Centric Control of Replicated Distributed Systems
Proponente(s): Hervé Paulino
Créditos: 42 ECTS
Área científica: Computer Systems and Networks
Início preferencial: Qualquer semestre
URL:
Já estão em curso trabalhos preliminares executados pelo alunos:
Breve descrição: The goal of this thesis is to extend data-centric concurrency (DCC) control to cope with replicated data systems, with the purpose of deducing the consistency model to be assigned to each operation accessing replicated data.

The approach is to shift all user-given information to data declaration and, with that, have two advantages: (i) centralize reasoning, making easier to reason as code scales, and (ii) automatically keep track of all accesses to replicated data, hence removing the burden from the programmer. The latter advantage comes directly from the application of the DCC model the replicated contexts, given that we already track these accesses for single replica environments.

Having all data accesses tracked, we will be able to deduce the side effects, and even refactor the operations so that they may use commutable versions of these side effects.
Observações: The thesis requires a prototype implementation in Java.
The work is be developed in the context of the DeDuCe research project.
There are scholarships available for students with good curriculum.