Data replication in smart grids: Using genetic algorithms to build online partitioning schemes

With the growth of alternative energy sources and the poorly evolved power delivery infrastructures, Smart Grids are emerging as a feasible alternative to adapt electric networks to current demands. In contra- position to classic electrical architectures, Smart Grids encompass a fully distributed scheme with several data generation sources. Current data storage and replication systems fail at both coping with such overwhelming amount of heterogeneous data and satisfying the stringent requirements posed by this technology: dynamic nature of the physical resources, continuous ow of information and autonomous behavior demands. The purpose of this paper is to take advantage of soft computing strategies to face these challenges and, thus, present a hybrid system that mixes data replication and partitioning policies with genetic algorithms and unsupervised learning by means of an online approach. Conducted experiments show that the proposed system outperforms previous proposals and truly ts with the Smart Grid premises.