Stabilizing Agreement is Impossible in Delayed Message Passing ModelsFebruary 14th, 2024
FiledFebruary 14th, 2024Most distributed computing research has focused on terminating problems like consensus and similar agree- ment problems. Non-terminating problems have been studied exhaustively in the context of self-stabilizing distributed algorithms, however, which may start from arbitrary initial states and can tolerate arbitrary tran- sient faults. Somehow in-between is the stabilizing consensus problem, where the processes start from a well-defined initial state but do not need to decide irrevocably and need to agree on a common value only eventually. Charron-Bost and Moran studied stabilizing consensus in synchronous dynamic networks con- trolled by a message adversary. They introduced the simple and elegant class of min-max algorithms, which allow to solve stabilizing consensus under every message adversary that (i) allows at least one process to reach all other processes infinitely often, and (ii) does so within a bounded (but unknown) number of rounds. Moreover, the authors proved that (i) is