Heterogeneous secure multi-party computation

Abstract

The increased processing power and storage capacity of inhome and mobile computing devices has motivated their inclusion in distributed and cloud computing systems. The resulting diverse environment creates a strong requirement for secure computations, which can be realised by Secure Multi-Party Computation (MPC). However, MPC most commonly assumes that parties performing the secure computation have the same characteristics and evenly distributes the computation load. In a heterogeneous environment, MPC using the same approach would result in poor performance. In this paper, we propose a mechanism for MPC share distribution in such an environment and present an analysis of the gain in robustness and the corresponding computational and communication complexity. Our results show that the uneven share distribution is a worthwhile approach in diverse computing systems. © 2012 IFIP International Federation for Information Processing.