Efficiency and Security Trade-Off in Supporting Range Queries on Encrypted Databases

The database-as-a-service (DAS) model is a newly emerging computing paradigm, where the DBMS functions are outsourced. It is desirable to store data on database servers in encrypted form to reduce security and privacy risks since the server may not be fully trusted. But this usually implies that one has to sacrifice functionality and efficiency for security. Several approaches have been proposed in recent literature for efficiently supporting queries on encrypted databases. These approaches differ from each other in how the index of attribute values is created. Random one-to-one mapping and order-preserving are two examples. In this paper we will adapt a prefix- preserving encryption scheme to create the index. Certainly, all these approaches look for a convenient trade-off between efficiency and security. In this paper we will discuss the security issues and efficiency of these approaches for supporting range queries on encrypted numeric data