Usually you have at least two replicas of data between the nodes of GlusteFS cluster. 

If you need details about handover between the nodes (in case of one node fail) you can check [this][1] answer in Unix and Linux SE site


  [1]: https://unix.stackexchange.com/questions/213705/glusterfs-how-to-failover-smartly-if-a-mounted-server-is-failed