impala-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Henry Robinson (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (IMPALA-5481) RowDescriptors should be shared, rather than copied
Date Wed, 21 Jun 2017 03:03:00 GMT

     [ https://issues.apache.org/jira/browse/IMPALA-5481?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Henry Robinson resolved IMPALA-5481.
------------------------------------
       Resolution: Fixed
    Fix Version/s: Impala 2.10.0

https://github.com/apache/incubator-impala/commit/317c413a00bd9b3b29eeaf2efe556c2e924e2d74

> RowDescriptors should be shared, rather than copied
> ---------------------------------------------------
>
>                 Key: IMPALA-5481
>                 URL: https://issues.apache.org/jira/browse/IMPALA-5481
>             Project: IMPALA
>          Issue Type: Improvement
>    Affects Versions: Impala 2.10.0
>            Reporter: Henry Robinson
>            Assignee: Henry Robinson
>             Fix For: Impala 2.10.0
>
>
> One of the {{RowBatch}} c'tors copies the row descriptor into the row batch. This leads
to a lot of allocation churn since {{RowDescriptor}} contains some vector members, and since
the descriptor is usually the same the copies are unnecessary.
> Instead, we should consider allocating the {{RowDescriptor}} once from an object pool,
and sharing it amongst all row batches that need that descriptor. 
> In some tests, {{RowDescriptor()}} shows up as 20% of the tcmalloc allocation time.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message