The article explores integrated approaches to design, management and security of critical information infrastructure in the context of the digital ecosystem of a megalopolis. A number of formalized models and methods are presented, including the use of entropy approach, Markov decision processes and structurally optimized neural networks. The central place in the study is occupied by the problem of integration of heterogeneous systems, for the solution of which it is proposed to use a service-oriented architecture and data buses. A criteria model for comparative analysis of integration methods at the design stage is developed. In addition, a proactive cybersecurity mechanism based on big data analysis is presented to detect anomalies and counteract false data injections, which is key to the sustainable functioning of a smart city.