This talk concerns integrative data analytics, with the emphasis of distributed inference in data integration. As data sharing from related studies become of interest, statistical methods for a joint analysis of all available datasets are needed in practice to achieve better statistical power and detect signals that are otherwise impossible to be captured from a single dataset alone. A major challenge arising from integrative data analytics pertains to principles of information aggregation, learning data heterogeneity, inference and algorithms for model fusion. Generalizing the classical theoretical foundation of information aggregation, we propose a new framework of distributed inference implemented by divide-and-conquer algorithms to handle massive large-scale data. I will focus on a new framework of renewable estimation, a generalization of Fisher’s Fiducial inference. I discuss both conceptual formulation and theoretical guarantees of the proposed method, and illustrate its performance via numerical examples.