Simulating Cheated Results Dissemination for Volunteer Computing

Abstract

In this paper, we present three novel distribution algorithms (Epoch Distribution Algorithm, Sliding Window Distribution Algorithm, and Extended Epoch Distribution Algorithm) in unstructured peer-to-peer networks for volunteer computing. In volunteer computing, the overall job is split into several sub jobs. These sub jobs are given to the (untrusted) PCs of volunteers. We show the performance of our algorithms with respect to double computed sub jobs. Furthermore, we present our idea of using cellular automata to simulate, evaluate, and visualize the cheated sub job result dissemination. Finally, we estimate and visualize the cheat detection behaviour of our cheat detection algorithm for volunteer computing.