[
https://issues.apache.org/jira/browse/HDFS-12443?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16224885#comment-16224885
]
Weiwei Yang edited comment on HDFS-12443 at 10/30/17 1:05 PM:
--------------------------------------------------------------
Hi [~linyiqun]
Please take a look at the pdf I just uploaded
[HDFS-12443-SCM-blockdeletion-throttle.pdf|https://issues.apache.org/jira/secure/attachment/12894736/HDFS-12443-SCM-blockdeletion-throttle.pdf],
the key issue here is we want to properly throttle the size of message sent to
DN, see chart page 1 {{CommandQueue}}. In every interval, we want each DN gets
a throttled number of {{DeletedBlocksCommand}} to process. But delLog has no
notion of DN, so it's a bit difficult to get there. Please let me know your
thought, thanks.
was (Author: cheersyang):
Hi [~linyiqun]
Please take a look at the pdf I just uploaded
^HDFS-12443-SCM-blockdeletion-throttle.pdf, the key issue here is we want to
properly throttle the size of message sent to DN, see chart page 1
{{CommandQueue}}. In every interval, we want each DN gets a throttled number of
{{DeletedBlocksCommand}} to process. But delLog has no notion of DN, so it's a
bit difficult to get there. Please let me know your thought, thanks.
> Ozone: Improve SCM block deletion throttling algorithm
> -------------------------------------------------------
>
> Key: HDFS-12443
> URL: https://issues.apache.org/jira/browse/HDFS-12443
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: ozone, scm
> Reporter: Weiwei Yang
> Assignee: Yiqun Lin
> Labels: OzonePostMerge
> Attachments: HDFS-12443-HDFS-7240.001.patch,
> HDFS-12443-HDFS-7240.002.patch, HDFS-12443-HDFS-7240.002.patch,
> HDFS-12443-SCM-blockdeletion-throttle.pdf
>
>
> Currently SCM scans delLog to send deletion transactions to datanode
> periodically, the throttling algorithm is simple, it scans at most
> {{BLOCK_DELETE_TX_PER_REQUEST_LIMIT}} (by default 50) at a time. This is
> non-optimal, worst case it might cache 50 TXs for 50 different DNs so each DN
> will only get 1 TX to proceed in an interval, this will make the deletion
> slow. An improvement to this is to make this throttling by datanode, e.g 50
> TXs per datanode per interval.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
