Abstract: In view of the traditional distributed data security algorithm under cloud computing, there has been a problem that the quality of data reading is not high, and the repetition rate is high. The range of attribute based cloud computing distributed data security read algorithm, using the empirical distribution function and kernel estimation of distribution function under distributed cloud computing data are estimated; then to transform the distribution function of distributed data assuming uniform distribution of inspection inspection; then distributed data analysis based on the results, the range of feature range type attribute, the distributed data domain into multiple sub intervals according to distribution of samples into the target database, complete cloud computing security read based on distributed data. The experimental results show that the proposed algorithm can improve the security of the distributed data reading speed and reading accuracy, reduce the computational complexity, and the application scope is wide.