According to the traditional sequential pattern mining algorithm with high complexity and not suitable for massive data, this paper puts forward Ha-GSP, which is a sequential pattern mining algorithm based on cloud computing platform Hadoop. On the basis of GSP algorithm, this paper researchs on implementation of GSP algorithm on Hadoop platform, and uses MapReduce model to solve the problem of massive data mining, and then finds frequent sequences. The experimental results show that: improved algorithm with less time, high efficiency.