Abstract:A new analytical method for improving the performance of a network attached optical jukebox is presented by means of artificial neural networks. Through analyzing operation (request) process in this system, the mathematics model and algorithm are built for this storage system, and then a classified method based on artificial neural networks for this system is proposed. Simulation results testified the feasibility and validity of the proposed method that it could overcome the drawbacks of the frequent I/O operation and provide an effective way for using the Network Attached Optical Jukebox.

Classifying Algorithm Based on A Fuzzy Neural Network for the Control of A Network Attached Optical Jukebox