With the advent of the era of cloud computing, the amount of application data increases dramatically, and personalized recommendation technology becomes more and more important. This paper mainly studies the collaborative filtering detection algorithm in the cloud computing environment. The algorithm migrates the collaborative filtering detection technology and applies it to the cloud computing environment. It shortens the recommendation time by using the advantages of clustering. A new recommendation algorithm can improve the accuracy of recommendation, and proposes a parallel collaborative filtering recommendation algorithm based on project. The algorithm is designed with programming model The experimental results show that the proposed algorithm has shorter running time and better scalability than the existing parallel algorithm.