Guerraoui, RachidPavlovic, Matej2015-06-232015-06-232015-06-232013https://infoscience.epfl.ch/handle/20.500.14299/115317This project aims at building a dynamic and scalable framework for Byzantine fault tolerant distributed computation. By organizing nodes of a distributed system in fault-tolerant clusters, we mask Byzantine failures of single nodes and provide an abstractions of reliable and highly available units at the level of these clusters.distributed systemsbyzantine fault tolerancetext::conference output::conference poster not in proceedings