In order to solve the problems of low efficiency of data input and retrieval, difficult storage and management of the unstructured data, poor scalability and weak ability to respond to multi-users real-time requests that the massive data of space object optical property faced, a hybrid storage strategy is proposed. Firstly, the data are dealt with and the retrieval strategy is constructed by using the combinatorial sequence regularity of the pitch and azimuth of the optical source and the detector. Secondly, combined with the requirements of data storage, processing and retrieval in the applications of space object optical property data, the efficient hybrid storage strategy is proposed, and storage architecture is designed. Finally, scenarios of simulated data input and retrieval are selected for the experimental verification. The experimental results show that the data input efficiency of the strategy proposed is more than 17 times higher than that of the traditional relational database in the case of storing only numerical data, and more than 34 times separately in the case of storing numerical data and image files. The data retrieval efficiency under multi-user requests is much higher than that of the traditional relational database. The results indicate that the hybrid storage strategy proposed can well meet the real-time data requirements of simulation, testing, and experiment related to the space object optical properties.