On Mon, May 23, 2011 at 9:59 AM, praveenesh kumar <[email protected]> wrote: > Please any suggestions..!! > > > ---------- Forwarded message ---------- > From: praveenesh kumar <[email protected]> > Date: Sun, May 22, 2011 at 2:23 PM > Subject: Hbase question,,!! > To: [email protected] > > > Okay guys.. so I have hadoop cluster of 5 nodes.. the configuration look > like this. > > 162.192.100.53 -- Master as well as slave > > Slave nodes : > > 162.192.100.52 > 162.192.100.51 > 162.192.100.50 > 162.192.100.49 > > Now I want to implement HBASE on my hadoop cluster.. What can be the best > configuration for my HBASE based on my hadoop structure ?? >
What are the choices? It's hard to say 'best' without knowing 'among these choices'. Normally, you install HBase, and configure it to use a directory in your HDFS as its root directory. Like datanodes in Hadoop, there are regionservers in HBase, and like the master, there is a HMaster. To ensure fault tolerance and consensus, a ZooKeeper instance also runs alongside. Is this what you wanted to know? Have you started with any documentation/tutorial, etc? -b
