[ https://issues.apache.org/jira/browse/HBASE-12233?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14172123#comment-14172123 ]
Francis Liu commented on HBASE-12233: ------------------------------------- Thanks for the feedback guys. Just to reiterate, we do have a real need to scale and we will improve/fix what's necessary be it on hbase, hdfs, or infra to meet that need. So far based on our experimentation and experience from actual large cluster deployments, it is hbase that is preventing us from scaling to our expected needs and splitting meta is the clear solution. {quote} We could try this just with a 0.98 based feature branch and determine if it's worth pursuing further and elsewhere. {quote} [~apurtell] Thanks for volunteering. Is this a feature branch used to determine whether it is worthwhile to bring back root (and split meta) or a feature branch to determine whether it is feasible to backport it to 0.98? What would be the measure of success? IMHO a 0.98 feature branch should be enough as a proving ground for either scenario. > Bring back root table > --------------------- > > Key: HBASE-12233 > URL: https://issues.apache.org/jira/browse/HBASE-12233 > Project: HBase > Issue Type: Sub-task > Reporter: Virag Kothari > Assignee: Virag Kothari > Attachments: HBASE-12233.patch > > > First step towards splitting meta. -- This message was sent by Atlassian JIRA (v6.3.4#6332)