AOC : Algorithms And Combinatorial Optimization

This axis focuses on optimization problems modeled using graphs and hypergraphs. Our goal is to provide fine-grained characterizations of their complexity, their solution spaces through polyhedral study, and their degree of approximability. The main results obtained involve: characterizing the complexity and approximation of fundamental or applied combinatorial optimization problems; studying the polyhedron induced by the set of feasible solutions of a problem; and finally, leveraging graph properties and graph algorithms to model and solve, in original ways, problems arising in large-scale or specific graphs.

This axis focuses on the resolution (exact or approximate) of combinatorial optimization problems through mathematical programming. Approaches used include constraint and column generation schemes, as well as decomposition and relaxation methods. We also address problem-solving through non-linear models, reoptimization methods, and matheuristic-based approaches.

This third axis focuses on distributed computing. Our research lies at the intersection of three fields: distributed algorithms, software, and architectures. We mainly address the following issues: middleware for execution environments and scientific workflows in cloud contexts, and for big data management; large-scale system modeling and the calculation of fault-tolerance properties; parallel and distributed numerical computing; application-level fault tolerance and irregular parallel applications; placement optimization for distributed shared virtual memory and cloud service optimization; and service-related challenges.