What is the Backup node in Hadoop?
Backup node maintain same namespace in memory as that of primary namenode, so it’s RAM requirement is same as Namenode.In Secondary namenode and checkpoint node, checkpoint is created by downloading fsimages and edit logs on their local file system from active primary namenode and merges these these edit logs files and new fsimgaes copy is saved in local file system. But unlike secondary namenode, backupnode does not need to download fsimage and edit logs from active primary namenode to create a checkpoint, as it already has an up-to-date copy of namespace in it’s own main memory. So creating a checkpoint in backup node is just to save a copy of namespace from main memory to local file system.
And why is this here?
Sorry but MapR doesn't have a SPOF of a name node.
Retrieving data ...