The invention, which belongs to the field of super-large-scale data storage management, relates to a hierarchical storage optimization method for super-large-scale drug data. The method comprises thefollowing steps: step one, constructing a cluster storage resource management system based on a distributed multistage storage structure, and allocating specific cluster storage resources to specificusers, user groups or jobs; step two, performing characterization processing on the jobs, dividing job categories, and intelligently scheduling the jobs to servers of data blocks required by the jobs;step three, designing a data classification model, mapping and storing massive result data generated in the computer-aided drug design process by applying the model, and segmenting the generated datainto data blocks to be respectively stored on servers at corresponding storage levels; and step four, designing a corresponding I / O method for each level of storage structure and the characteristic attributes thereof, dynamically scheduling I / O requests, and optimizing the I / O scheduling strategy of each level of storage structure. According to the method, the I / O performance in the supercomputing environment is improved, and heterogeneous storage and platform development and utilization of super-large-scale drug data are realized.