hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Suresh Srinivas (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-395) DFS Scalability: Incremental block reports
Date Fri, 22 Jul 2011 22:00:10 GMT

    [ https://issues.apache.org/jira/browse/HDFS-395?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13069805#comment-13069805

Suresh Srinivas commented on HDFS-395:

Tomasz, cool stuff.

Some comments:
# ReceivedDeletedBlockInfo.java is missing in the patch.
# DataNodeProtocol.java
#* Please update the DataNodeProtocol#blockReceivedAndDeleted() javadoc to reflect the new
# DataNode.java
#* Minor: DataNode.java indendtation of receivedAndDeletedBlockList.wait()
#* I am not clear on why you are setting deleteReportInterval to DFS_BLOCKREPORT_INTERVAL_MSEC_KEY
and blockReportInterval to 2*DFS_BLOCKREPORT_INTERVAL_MSEC_KEY. Can you retain blockReportInterval
same as before and use some suitable value for deleteReportInterval. This could be 100 times
heartbeat. The reason to be aggressive with this is, NN is going to keep the replica in a
data structure, until delete ack. I know that it is piggy backed with blockReceived; but still
see no bad side to sending it more frequently.
#* reportReceivedDeletedBlocks() - save the pendingReceivedRequests in currentReceived in
side synchronixed(receivedAndDeletedBlockList)?
#* pendingReceivedRequests accessed without lock in OfferService - you can move this check
into reportReceivedDeletedBlocks()

> DFS Scalability: Incremental block reports
> ------------------------------------------
>                 Key: HDFS-395
>                 URL: https://issues.apache.org/jira/browse/HDFS-395
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: dhruba borthakur
>            Assignee: Tomasz Nykiel
>         Attachments: blockReportPeriod.patch, explicitDeleteAcks.patch
> I have a cluster that has 1800 datanodes. Each datanode has around 50000 blocks and sends
a block report to the namenode once every hour. This means that the namenode processes a block
report once every 2 seconds. Each block report contains all blocks that the datanode currently
hosts. This makes the namenode compare a huge number of blocks that practically remains the
same between two consecutive reports. This wastes CPU on the namenode.
> The problem becomes worse when the number of datanodes increases.
> One proposal is to make succeeding block reports (after a successful send of a full block
report) be incremental. This will make the namenode process only those blocks that were added/deleted
in the last period.

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message