Research On Android Based Cache-prefetch Algorithms For Mobile Cloud Storage

With rapid development of mobile Internet technology, the limited storage space of the mobile terminal is difficult to meet the needs of users, and the cloud storage technology provides a solution for this demand. Compared with the wired network, there are still some problems like high access latency and unstable network connection to use cloud storage services during mobile Internet. To resolve these problems, we can use caching technology to improve the efficiency of visiting cloud storage servers by mobile terminal equipment, and save the user's network overhead.For low hit ratio caused by existing cache prefetching algorithms, we propose a dynamic window prefetching strategy based on list order and a hierarchical cache replacement algorithm. The prefetching strategy take advantage of the feature that users browse the files on a mobile device will usually slide in order, through subsequent access probability calculation to select a file to local in advance prefetching. Through the LRU and LFU and other typical cache replacement algorithm to analyze and improve, a new hierarchical cache replacement algorithm is to be designed and implemented based on NAND flash memory, considering the access time and access frequency.The prefetching strategy and the replacement algorithm are verified, tested and analyzed on the Android mobile device. The Experimental results showed that the system can save nearly 40% of the access time when compared to non-prefetch system, and enhance the hit rate to 0.8, so these showed that the cache and prefetching technology can effectively improve the hit rate in mobile cloud storage applications,and reduce access time delay.