Hardening Functions for Large Scale Distributed Computations

Many recent large-scale distributed computing applications utilize spare processor cycles of personal computers that are connected to the Internet. The resulting distributed computing platforms provide computational power that previously was available only through the use of expensive supercomputers. However, distributed computations running in untrusted environments raise a number of security concerns, including the potential for disrupting computations and for claiming credit for computing that has not been completed (i.e., cheating). This paper presents two strategies for hardening selected applications that utilize such distributed computations. Specifically, we show that carefully seeding certain tasks with precomputed data can significantly increase resistance to cheating and to disrupting the computation. We obtain similar results for sequential tasks by sharing the computation of tasks among nodes. In each case, the associated cost is significantly less than the cost of assigning tasks redundantly.