NameNode does not need to run a replicator to choose a random DataNode
----------------------------------------------------------------------
Key: HADOOP-5595
URL: https://issues.apache.org/jira/browse/HADOOP-5595
Project: Hadoop Core
Issue Type: Improvement
Components: dfs
Reporter: Hairong Kuang
Assignee: Hairong Kuang
Fix For: 0.21.0
Attachments: randomDataNode.patch
FSNamesystem#getRandomDatanode uses a replicator to choose a random DataNode.
It's an overkill. Instead, Topology#chooseRandom is a much light-weight way to
choose a random DataNode from a cluster map.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
