A system and method for failover in a distributed system may comprise a computing device that receives information associating computing nodes with ordinal identifiers, such that the computing nodes are divided into at least a first and second subset based on the identifiers. The identifiers may further define an ordering of the subsets. When failover occurs, candidate computing nodes may be identified and selected based at least in part on the ordering. Secondary considerations, including functions performed by other members of a candidate's subset, are considered when identifying candidate nodes.