[ https://issues.apache.org/jira/browse/HDFS-3689?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14104243#comment-14104243 ]
Todd Lipcon commented on HDFS-3689: ----------------------------------- Nice idea, Andrey. Jing, can you explain more what the use case is for padding blocks in Hive? If we had a new API like {{DFSOutputStream.writeZeros(int length);}}, we could implement it with a special packet type, and have the DN just call "truncate" to extend the size of the underlying block file. It would have to write the appropriate checksum data as well, but that's still a ~99% reduction in IO. As far as I'm aware, all commonly used local file systems do support sparse files. > Add support for variable length block > ------------------------------------- > > Key: HDFS-3689 > URL: https://issues.apache.org/jira/browse/HDFS-3689 > Project: Hadoop HDFS > Issue Type: New Feature > Components: datanode, hdfs-client, namenode > Affects Versions: 3.0.0 > Reporter: Suresh Srinivas > Assignee: Suresh Srinivas > Attachments: HDFS-3689.000.patch, HDFS-3689.001.patch > > > Currently HDFS supports fixed length blocks. Supporting variable length block > will allow new use cases and features to be built on top of HDFS. -- This message was sent by Atlassian JIRA (v6.2#6252)