Social Network services have quickly become a powerful means by which people share real-time messages. Typically, social networks are modeled as large underlying graphs. Responding to this emerging trend, it becomes critically important to interactively view and analyze this massive amount of data from different perspectives and with multiple granularities. While Online analytical processing (OLAP) is a powerful primitive for structured data analysis, it faces major challenges in manipulating this complex interconnecting data. In this paper, we suggest a new data warehousing model, namely Social Graph Cube to support OLAP technologies on multidimensional social networks. Based on the proposed model we represent data as heterogeneous information graphs for more comprehensive illustration than the traditional OLAP technology. Going beyond traditional OLAP operations, Social Graph Cube proposes a new method that combines data mining area and OLAP operators to navigate through dimension hierarchies. Experimental results show the effectiveness of Social Graph Cube for decision-making.