hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Walter Su (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-8719) Erasure Coding: client generates too many small packets when writing parity data
Date Wed, 08 Jul 2015 07:57:05 GMT

    [ https://issues.apache.org/jira/browse/HDFS-8719?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14618170#comment-14618170
] 

Walter Su commented on HDFS-8719:
---------------------------------

bq. If the last packet is not full, it will be enqueued in closeImpl()
I debug TestDFSStripedOutputStream with filesize = blockSize*dataBlocks*3 -1

You're right it will be enqueued in closeImpl(). But the stack trace shows:
closeImpl()
--> flushBuffer()
--> writeChunk()
--> if (currentPacket.getNumChunks() == currentPacket.getMaxChunks().) is true
--> enqueueCurrentPacketFull()
--> remainingBytes=1

> Erasure Coding: client generates too many small packets when writing parity data
> --------------------------------------------------------------------------------
>
>                 Key: HDFS-8719
>                 URL: https://issues.apache.org/jira/browse/HDFS-8719
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Li Bo
>            Assignee: Li Bo
>         Attachments: HDFS-8719-001.patch, HDFS-8719-HDFS-7285-001.patch
>
>
> Typically a packet is about 64K, but when writing parity data, many small packets with
size 512 bytes are generated. This may slow the write speed and increase the network IO.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message