Graphs have been widely used for representing the relationship among users in online social networks (OSNs). As crawling an entire OSN is resource- and time-consuming, most of the existing work only pick a sampled subgraph for study. However, this may introduce serious inaccuracy into the analytic results, not to mention that some important metrics cannot even be calculated. In this work, we crawl the entire social network of Swarm, a leading mobile social app with more than 60 million users, using a distributed approach. Based on the crawled massive user data, we conduct a data-driven study to get a comprehensive picture of the entire Swarm social network. Our study provides a deep analysis of social interactions between Swarm users, and reveals the relationship between social connectivity and check-in activities.