The smart city operation and management center with a hierarchical data-driven architecture has already become one of the most widely used solutions for smart cities in practice, solving the problems associated with data acquisition, data gathering and storage, data processing, and data application. At present, the construction of smart city operation and management center faces bottlenecks such as incomplete top-level design theory, the insufficient integration capability of software and hardware, the low efficiency of data collection and aggregation, and the lack of intelligence in data analysis and application. Aiming to address the above problems, this paper proposes a 'two-dimension, three-layer, and six-goal' top-level design model for a smart city, with six principles for a smart city operational pattern, and focuses on three key technologies: (1) infrastructure integration and application, (2) multidimensional perception data collection and aggregation, and (3) intelligent data analysis and data service. Following the guidance of this model, Longgang District of Shenzhen has constructed a smart city operation and management center including integrated ICT infrastructure, as urban fine management system, and an intelligent urban data analysis and service system. Practice shows that the top-level design model of a smart city proposed in this paper has achieved successful results, and it thereby offers a model of a smart city that can be referenced and replicated.INDEX TERMS Data-driven, urban governance, smart city, top-level design.