[ https://issues.apache.org/jira/browse/HADOOP-6393?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Tom White resolved HADOOP-6393. ------------------------------- Resolution: Not A Problem I realized that this is actually already supported as long as you launch the master separately: {code} hadoop-ec2 launch-master --instance-type=c1.xlarge my-hadoop-cluster hadoop-ec2 launch-slaves my-hadoop-cluster 10 # use the instance type set in the configuration {code} > Allow master to run on larger EC2 instance > ------------------------------------------ > > Key: HADOOP-6393 > URL: https://issues.apache.org/jira/browse/HADOOP-6393 > Project: Hadoop Common > Issue Type: New Feature > Components: contrib/cloud > Reporter: Tom White > > The namenode often needs more memory, so it would be useful to allow it to be > run on a different instance size. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.