Common knowledge and consistent simultaneous coordination

摘要: There is a very close relationship between common knowledge and simultaneity in synchronous distributed systems. The analysis of several well-known problems terms has led to round-optimal protocols for these problems, including Reliable Broadcast, Distributed Consensus, the Firing Squad problem. These require that correct processors coordinate their actions some way but place no restrictions on behaviour faulty processors. In systems with benign processor failures, however, it reasonable be consistent those processors, assuming performs any action at all. We consider requiring consistent, simultaneous coordination. then analyze failure models. stronger requires definition knowledge, we study two definitions. many cases, definitions are actually equivalent, simple modifications previous solutions yield problems. When differ, show such cannot solved, even failure-free executions.