I don't think data node needs much memory, as Stu suggested. There will be requirement for memory when running a map reduce job. In that case, more memory is better for in memory sorting (on map) and in-memory copy (on reduce).
Namenode needs memory to hold meta data though. On Thu, Mar 10, 2011 at 11:32 AM, Rita <rmorgan...@gmail.com> wrote: > No, just a datanode. Nothing else. > > > > On Wed, Mar 9, 2011 at 11:30 AM, <stu24m...@yahoo.com> wrote: > >> Is anything else running on the datanodes? Datanodes themselves don't need >> too much memory. >> >> Take care, >> -stu >> ------------------------------ >> *From: * Rita <rmorgan...@gmail.com> >> *Date: *Wed, 9 Mar 2011 07:24:26 -0500 >> *To: *<hdfs-user@hadoop.apache.org> >> *ReplyTo: * hdfs-user@hadoop.apache.org >> *Subject: *datanode memory requirement >> >> All of my data nodes have 16GB of memory. Would increasing my Java virtual >> memory help in performance? >> >> -- >> --- Get your facts first, then you can distort them as you please.-- >> > > > > -- > --- Get your facts first, then you can distort them as you please.-- >
