[jira] [Updated] (HDFS-6268) Better sorting in NetworkTopology#pseudoSortByDistance when no local node is found

["Andrew Wang (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HDFS-6268?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Andrew Wang updated HDFS-6268:
------------------------------

    Attachment: hdfs-6268-3.patch

Here's a new patch which does the rework ATM proposed. This has the nice property of also code sharing with NetworkTopologyWithNodeGroup.

I realized we can't integrate the decom/stale sorting since NetworkTopology is in hadoop-common, but I don't think that's a biggie.

> Better sorting in NetworkTopology#pseudoSortByDistance when no local node is found
> ----------------------------------------------------------------------------------
>
>                 Key: HDFS-6268
>                 URL: https://issues.apache.org/jira/browse/HDFS-6268
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>    Affects Versions: 2.4.0
>            Reporter: Andrew Wang
>            Assignee: Andrew Wang
>            Priority: Minor
>         Attachments: hdfs-6268-1.patch, hdfs-6268-2.patch, hdfs-6268-3.patch
>
>
> In NetworkTopology#pseudoSortByDistance, if no local node is found, it will always place the first rack local node in the list in front.
> This became an issue when a dataset was loaded from a single datanode. This datanode ended up being the first replica for all the blocks in the dataset. When running an Impala query, the non-local reads when reading past a block boundary were all hitting this node, meaning massive load skew.



--
This message was sent by Atlassian JIRA
(v6.2#6252)