Section: New Results

Secure Global Computing on Asymmetric Architecture

Participants : Benjamin Nguyen [correspondent] , Philippe Pucheral, Quoc-Cuong To.

Current applications, from complex sensor systems (e.g. quantified self) to online e-markets acquire vast quantities of personal information which usually ends-up on central servers. Decentralized architectures, devised to help individuals keep full control of their data, hinder global treatments and queries, impeding the development of services of great interest. In this study, we promote the idea of pushing the security to the edges of applications, through the use of secure hardware devices controlling the data at the place of their acquisition. To solve this problem, we propose secure distributed querying protocols based on the use of a tangible physical element of trust, reestablishing the capacity to perform global computations without revealing any sensitive information to central servers. This leads to execute global treatments on an asymmetric architecture, composed of a powerful, available and untrusted computing infrastructure (server or cloud), and a large set of low powered, highly disconnected trusted devices. Given our large scale data centric applications (e.g. nationwide surveys), we discard solutions based on secure multi-party computation, which do not scale. We have studied two different computing paradigms on this architecture: our first contribution was to study the execution of Privacy Preserving Data Publishing (PPDP) algorithms on such an architecture, and provided generic protocols to deal with all kinds of PPDP algorithms, which are robust against honest-but-curious and malicious adversaries [12] , including vulgarization aspects [25] . Our second contribution was to study general SQL queries in this same execution context. For now, we have concentrated on the subset of SQL queries without joins, but including Group By and aggregates, and show how to secure their execution in the presence of honest-but-curious attackers [19] . Cost models and experiments demonstrate that this approach can scale to nationwide infrastructures [20] [16] . This work is part of Quoc-Cuong To's Ph.D. thesis started in sept. 2012, and should be extended in particular to cover joins. We also plan to extend this general framework through a collaboration with INSA Centre Val de Loire, LIFO Lab and University of Paris Nord, LIPN lab, to study the secure execution of Map/Reduce on the Asymmetric Architecture.