Research Of Cache Strategy Based On Social Relationship In The Environment Of Mobile Cloud Computing

In recent years, mobile applications grew rapidly with the development of Android and iOS platforms. Mobile cloud computing are being used more frequently in development, that is more and more mobile application are using the architecture of "Cloud+Client" to reduce the cost. Most of applications on these smart phones are based on users, and these data are always generated by users. When users want relative data, it is not very realistic to request every time. So a suitable cache technology is required. Traditional cache technologies pay more attention to data access frequency or last access time, but do not consider more about data generators’ relationship with others. Data analysis shows that people spend most of time when using mobile applications. That is why social element is a basic component in most mobile applications.In the environment of mobile cloud computing where exists social relations, data access is closely related to users’relationships, so this factor should be suitable used in cache technologies.Firstly, we analyzed how data request in common mobile applications which developed in mobile cloud computing environment. By abstracting data’s feature in this environment, we definite reasonable cache object model.Secondly, we described how to compute cache objects’value in detail. We combined users’relationships with the classic cache algorithm LRU. Not only considered the access times of each data, but also users’relationships by computing the closeness value between the data requestor and the generator.After that this paper explained the process of this cache replacement algorithm.In the last, this paper designed a reasonable experiment to testify the algorithm combined with the exhibition project. Because of the data request from mobile client conform to Zipf s law, we simulated the data request queue to make it similar to the real. Then we compare the results of hit rate computed by LRU and SAIU. The result shows that in the environment of mobile cloud computing where exists social relations, our strategy performs better.