ABSTRACT:

Computational Private Info Retrieval (cPIR) protocols permit a client to retrieve one bit from a database, while not the server inferring any information regarding the queried bit. These protocols are too pricey in practice as a result of they invoke complicated arithmetic operations for every little bit of the database. In this paper, we gift pCloud, a distributed system that constitutes the first try toward practical cPIR. Our approach assumes a disk-primarily based design that retrieves one page with a single question. Using a striping technique, we distribute the database to a range of cooperative peers, and leverage their computational resources to method cPIR queries in parallel. We implemented pCloud on the PlanetLab network, and experimented extensively with many system parameters. Our results indicate that pCloud reduces considerably the query response time compared to the ancient client/server model, and has a very low communication overhead. Additionally, it scales well with an increasing range of peers, achieving a linear speedup.