With increased improvement on the capability and performance of software systems, enterprises have improved the management efficiency and enhanced the business model. Meanwhile, as software systems become more and more complex, severe challenges for the management of software systems are encountered. This paper presents a method for measuring and predicting software system state based on hidden Markov model. It establishes the linkage between the exterior state (the observation state) and the interior state (the hidden state) of the software system. K-means method is applied to construct the observation state of system. Triple order exponential smoothing is used to predict the future state of the system exterior state which changes cyclically. The experimental analysis on B/S information management system shows that the proposed method has high accuracy for measuring and predicting the software system state.