In this paper, we tackle the joint optimization of the network topology and the optimal location of distributed renewable energy resources in a Microgrid (MG). The MG network topology optimization problem is focused on obtaining network deployments with minimal cost, whereas the location of distributed renewable generation is associated with the minimization of the electricity losses in the MG lines. In order to solve this joint optimization problem, we analyze the efficiency of the Harmony Search (HS), a novel meta-heuristic solver inspired by the music improvisation procedure observed in jazz bands. We consider two different approaches, the first one is a single-objective formulation of the problem, where the classical HS is applied with some adaptations. The second approach is to consider a multi-objective version of the HS algorithm, able to evolve a whole family of solutions in a Pareto front. Both approaches have been tested on two small-sized MGs: an 8 node MG and a 12 node MG, and results have been compared to an 8 node and a 12 node baseline scenario, respectively, obtaining improvements of up to 42%.