Storage efficient and unconditionally secure private information retrieval

Abstract

A method of storing and retrieving a set of original data (E1,...,En) in and from a plurality of remote servers (SP1,..., SPI+1), comprises a coding step which consists in creating a set of coded data (S1,...,SN) from the set of original data (E1,...,En), a storing step which consists in storing the set of coded data (S1,...,SN ) into the plurality of remote servers (SP1,..., SPI+1). Each server (SP1,..., SPI+1) of the said plurality of servers stores only a respective part of the set of coded data (S1,...,SN) and the method comprises a step which consists in generating a table (T1,T2) which indicates which respective part of the set of coded data (S1,...,SN ) is stored in which server of the said plurality of remote servers (SP1,..., SPI+1).

Description

technical field [0001] The present invention relates to storing information in a data storage server, making queries to databases and securing such queries. Background technique [0002] Many files and databases are now stored on outsourced data storage servers. This trend is further intensified with the cloudification of many services and information systems of companies. While such outsourced databases are inexpensive to maintain and efficient in retrieving up-to-date information, they can pose significant risks in terms of security or confidentiality. Some solutions have been proposed to secure outsourced data stored in multiple clouds, but they do not address the confidentiality issue. In fact by accessing the database, the entity storing the database learns the queryer's or user's query, which may be a problem if the user's intent is to be kept secret. For example, an investor who queries a stock market database to obtain the current market value of certain stocks ma...

AI Technical Summary

Problems solved by technology

However, the current solution requires many replicated

Images