Mailing-List: contact mapreduce-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Mon, 24 Oct 2016 14:40:00 +0000 (UTC)
From: "Steve Loughran (JIRA)" <jira@apache.org>
To: mapreduce-issues@hadoop.apache.org
Message-ID: <JIRA.13014746.1477319987000.67639.1477320000322@Atlassian.JIRA>
In-Reply-To: <JIRA.13014746.1477319987000@Atlassian.JIRA>
References: <JIRA.13014746.1477319987000@Atlassian.JIRA> <JIRA.13014746.1477319987588@arcas>
Subject: [jira] [Created] (MAPREDUCE-6800)
 FileInputFormat.singleThreadedListStatus to use listFiles(recursive)
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Mon, 24 Oct 2016 15:07:03 -0000

Steve Loughran created MAPREDUCE-6800:
-----------------------------------------

             Summary: FileInputFormat.singleThreadedListStatus to use listFiles(recursive)
                 Key: MAPREDUCE-6800
                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-6800
             Project: Hadoop Map/Reduce
          Issue Type: Improvement
          Components: mrv2
    Affects Versions: 2.7.3
            Reporter: Steve Loughran
            Priority: Minor


{{FileInputFormat.singleThreadedListStatus}} does recursive directory walks to pick files to scan. This is very inefficient on object stores, and can be bypassed if {{listFiles(recursive=true)}} can be used instead.

Based on the experience of SPARK-2984, it should also be resilient to a source file going away during the iteration, downgrading an FNFE to a "skip that nonexistent path"


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

---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: mapreduce-issues-help@hadoop.apache.org