Distributed Adaptive Metaheuristics Selection (DAMS) is a framework dedicated to adaptive optimization in distributed environments. We investigate the design of adaptive strategies allowing to control the local selection of metaheuristics and to coordinate their local executions with the aim of maximizing the performance of the whole distributed system. Inspired by the multi-armed bandit framework, we propose two distributed strategies. Our experimental analysis is performed on the simple oneMax problem for which the best metaheuristics that should be executed is known.