On the practical feasibility of secure distributed computing: a case study

Ref: Gregory Neven, Frank Piessens, and Bart De Decker.In S. Qing and J. Eloff, editors, Information Security for Global Information Infrastructures, IFIP TC11 Fifteenth Annual Working Conference on Information Security, volume 175 of IFIP Conference Proceedings, pages 361-370. Kluwer Academic Publishers, 2000.

Abstract: Secure Distributed Computing addresses the problem of performing a computation with a number of mutually distrustful participants, in such a way that each of the participants has only limited access to the information needed for doing the computation. Over the past decade, a number of solutions for this problem have been developed. The various proposed solutions differ in the cryptographic primitives that are used, and in the class of computations that can be performed. However, all sufficiently general solutions have one thing in common: the communication overhead between the involved parties seems to be prohibitive.

In this paper, we consider a concrete instance (with considerable practical interest) of the general problem of secure distributed computing, and we investigate how bad the communication overhead really is. This involves tailoring the different general solutions to the specific problem at hand, optimizing them for minimal communication overhead, and evaluating the resulting communication overhead.

Our conclusion is that the communication overhead is still large, but can be substantially reduced using specific techniques. Moreover, agent-techology makes Secure Distributed Computing more appealing,
since the mobile agents can position themselves close to each other.

PS | PDF | Powerpoint presentation