Cloud Architecture for Effective Cost towards Differential Query Services

Cloud computing as a developing technology trend is estimated to reshape the advances in info technology. In a cost-efficient cloud infrastructure, a user can tolerate a certain degree of interval while retrieving info from the cloud to decrease costs. In this paper, we address two major issues in such an environment: secrecy and efficiency. We first review a secretive keyword based file recovery scheme that was formerly proposed by Ostrovsky. Their structure allows a user to get files of interest from a not trusted server without disclosing any info. The main disadvantage is that it will cause a huge querying above incurred on the cloud, and thus goes opposite the novel intention of cost efficiency. In this paper, we present a scheme, termed efficient information retrieval for ranked query, and based on an aggregation and distribution layer, to decrease querying above incurred on the cloud. In efficient information retrieval for ranked query, queries are classified into many ranks, where a higher ranked query can get a higher ratio of matched files. A user can get files on demand by selecting queries of dissimilar ranks. Large number of matched files is useful in this feature, but the user only wants a small subset of them. Under dissimilar parameter settings, extensive estimations have been conducted together analytical models and on a real cloud infrastructure, in order to examine the usefulness of our schemes.