Wind extinguishes a candle and energizes fire. Likewise with randomness, uncertainty, chaos: you want to use them, not hide from them. You want to be the fire and wish for the wind.
The fast probabilistic consensus (FPC), introduced recently by Serguei Popov (paper) terminates in most of the times in a minimal number of steps and is much faster than the theoretical upper bounds given by Popov. It is claimed that an additional layer of randomness allows the protocol to still work in critical situations where other protocols fail. These simulations show that the FPC is indeed very robust under various attack scenarios ->.
Introduction and set-up
We perform a first simulation-based study on the fast probabilistic consensus proposed by Serguei Popov (paper). Since the protocol is very safe and fast in non-critical situations, this study is mainly about attack strategies that try to reach disagreement of the honest nodes and considers mostly critical situations. The post is not intended to be a thorough scientific study of the FPC; it justs collects various observations in order to illustrate its functionalities.
How can a distributed network find consensus on the value of a bit?