Hi Oussama. 1) That's all the goal of Hadoop and HBase ;) You might want to ready Hadoop the Definitive guide and HBase the Definitive guide... 2) HBase is based on Hadoop and take advantage of it's repplication process. 3) There is also a way to backup the data manually or to configure replication between different clusters. 4) There is no need to copy between nodes. Backup is more between an external server and the cluster, or between 2 clusters.
HTH. JM 2013/8/12 Oussama Jilal <[email protected]> > Hello everyone, > > I have some questions that I wish to get answers to regarding how HBase > and HDFS manages the data nodes. > > Q1- Can I remove a node from a cluster without loosing data ? > Q2- If yes (Q1), does that depend on the replication of data between nodes > or I don't need to worry about it even if data is not replicated ? > Q3- If no (Q1), is there a way to copy the data manually ? > Q4- If yes (Q3), can I copy to another node that contains data or does it > have to be a new empty node ? > > That is all I can think of right now ... any help is very appreciated, > thank you. >
