This is a model that is useful to describe systems that have nodes that can crash and recover later on.
- The system is a set of processes .
- processes communicate by sending messages using TCP connections.
- process has a persistent storage.
- processes have two states: up and down.
- process may crash and recover indefinitely many times.
Reference: ZooKeeper’s atomic broadcast protocol: Theory and practice, Andr´e Medeiros, http://www.tcs.hut.fi/Studies/T-79.5001/reports/2012-deSouzaMedeiros.pdf