Recently, Mobile Internet became the most popular network in China. There are a lot of emerging applications growing with the development of Mobile Internet. This paper illustrate the design of a Hadoop-based tool, Flow Logs Analysis System (FLAS), for analysis these applications. FLAS can storage, process and compute the massive Mobile Internet flow logs efficiently. Furthermore, the traffic of leading Micro-blog(LM-blog) and e-business operators(E-shop) was analyzed using FLAS based on the flow logs collected with a passive network monitoring equipment placed on a large Internet Service Provider (ISP) in Southern China. And the distribution features of LM-blog and E-shop traffic, as well as the flow duration and the relationship between packets and bytes were investigated.