As the core equipment of the storage system, a hard disk may cause data loss, system crash, and even business interruption if it fails. In order to improve the reliability and security of data centers, more and more deep learning methods have emerged to predict the remaining useful life (RUL) of hard disk drives (HDDs), but these methods have the problem of poor long-term failure prediction due to ignoring the time information in the data. In order to solve these problems, we propose a method for predicting the remaining useful life of HDD based on bidirectional LSTM and Transformer. The data is first preliminarily extracted by bidirectional LSTM, and then the features are input to the Transformer for further optimization, and finally the remaining service life of the HDD is predicted according to the features. Experiments have shown that the proposed method is very effective and achieves better performance compared to many state-of-the-art methods.