Achieving Flow-Oriented Reliable Services in Cloud Computing

In cloud computing, a data owner outsources a huge dataset to cloud servers, and authorizes many clients to retrieve data of interest anytime and anywhere. Data outsourcing is helpful to lessen the burden on local data storage and also to improve quality of services. However, the cloud service provider (CSP) outside the data owner's trusted domain may return wrong query results to the client deliberately or inadvertently.Therefore, query authentication and query correction are of crucial importance to achieve reliable cloud services. In this paper,we consider a flow-oriented cloud environment, where query results will be transmitted in the form of data flow to the clients via networks. Specifically, we propose a two-phase query (TPQ) protocol to support efficient authentication and correction of multi-dimensional range queries. By ingeniously integrating signature chain with Reed Solomon codes, our proposed protocol allows a client to efficiently verify the correctness of query results while performing error correction in an adaptive way. Extensive experiments on a real dataset demonstrate effectiveness of our proposed protocol.