A New Method of Pefetching I/O Requests

Prefetching is one of the most important method to improve storage system performance. Without knowing any I/O semantics in device layer, it is not easy now for storage system to exploit semantic information and then to prefetch the data. Many prefetching policies have to relay on simple patterns such as sequentially, temporal locality and loop references to improve storage system performance. Therefore, according to characteristic of storage system, this paper not only introduces a new sequence degree-based clustering algorithm to find the storage areas which will be accessed frequently, but also adopts ARMA time series model to forecast the storage areas on which data will be read frequently later, and their corresponding request time. Moreover, to improve the forecast accuracy, this paper adopts dynamic parameter estimation policy to ARMA model. The results of a large number of simulations validate the accuracy of the clustering algorithm and the preciseness of the ARMA time series model of dynamic parameter estimation policy, and indicate that efficiency of cache prefetching can be greatly improved through applying the clustering algorithm and ARMA time series model.