[ https://issues.apache.org/jira/browse/HDFS-11814?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16117009#comment-16117009 ]
Andrew Wang commented on HDFS-11814: ------------------------------------ Hi Wei, thanks for working on this. Did you test with a highly concurrent workload? One concern is that performance will degrade with a lot of seeks. It would be good to look at disk statistics to make sure disk throughput is fully saturated during the run. > Benchmark and tune for prefered default cell size > ------------------------------------------------- > > Key: HDFS-11814 > URL: https://issues.apache.org/jira/browse/HDFS-11814 > Project: Hadoop HDFS > Issue Type: Improvement > Reporter: SammiChen > Assignee: Wei Zhou > Labels: hdfs-ec-3.0-must-do > Attachments: RS-Read.png, RS-Write.png > > > Doing some benchmarking to see which cell size is more desirable, other than > current 64k -- This message was sent by Atlassian JIRA (v6.4.14#64029) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org