A proposal for a reduced client workload model for querying encrypted databases in cloud

Abstract

Cloud computing offers a considerable number of advantages to clients and organizations that use several capabilities to store sensitive data, interact with applications, or use technology infrastructure to perform daily activities. The development of new models in cloud computing brings with it a series of elements that must be considered by companies, particularly when the sensitive data needs to be protected. In this research, a model that uses a trusted third party (TTP) to enforce database security in the cloud is proposed. First, the TTP performs a partition process over the data by using an index from one of the attributes in the table, and the TTP sends to the cloud the records in encrypted format with an index. Second, the TTP analyzes the client query to retrieve a segment of the data from the cloud-based on the query conditions. The final result is submitted to the client in which a minimum workload is executed. Some simulations were performed to evaluate the efficiency of the model by using two partition techniques: Histogram based, and Mondrian or Bisection Tree-based. The strategy of the model is to process as much of the work at the TTP site and securely transmit the result.