[ https://issues.apache.org/jira/browse/HDFS-6934?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14158863#comment-14158863 ]
Arpit Agarwal commented on HDFS-6934: ------------------------------------- Hi Nicholas, the proposed approach of computing checksum in DN when write to memory is requested sounds fine to me. The truth table also looks good. Perhaps the only concern could be that the CPU usage will show up in the DN process and not the client process, but the overall CPU usage of the node will not be affected as long as we ensure memory writes are always local. I filed HDFS-7192 for that, will fix it later. > Move checksum computation off the hot path when writing to RAM disk > ------------------------------------------------------------------- > > Key: HDFS-6934 > URL: https://issues.apache.org/jira/browse/HDFS-6934 > Project: Hadoop HDFS > Issue Type: Sub-task > Components: datanode > Reporter: Arpit Agarwal > Assignee: Tsz Wo Nicholas Sze > > Since local RAM is considered reliable we can avoid writing checksums on the > hot path when replicas are being written to a local RAM disk. > The checksum can be computed by the lazy writer when moving replicas to disk. -- This message was sent by Atlassian JIRA (v6.3.4#6332)