The “smart city” concept is becoming more and more popular for several reasons. Among them is the common understanding that urbanization should be sustainable in terms of quality of life for the citizens, financial viability, protection of the environment, etc. Key enablers for realizing this concept are the Internet of Things, M2M communications, and efficient and intelligent data management methods. In this presentation we describe experimentation with a large dataset of parking data in one of the largest smart city testbeds in the world: Santander. The work was performed in the context of the SmartSantander project and resulted in a novel parking prediction service.