Mailing-List: contact issues-help@spark.apache.org; run by ezmlm
Precedence: bulk
Date: Fri, 3 Nov 2017 15:33:00 +0000 (UTC)
From: "Bryan Cutler (JIRA)" <jira@apache.org>
To: issues@spark.apache.org
Message-ID: <JIRA.13105491.1506527887000.143475.1509723180719@Atlassian.JIRA>
In-Reply-To: <JIRA.13105491.1506527887000@Atlassian.JIRA>
References: <JIRA.13105491.1506527887000@Atlassian.JIRA> <JIRA.13105491.1506527887400@jira-lw-us.apache.org>
Subject: [jira] [Commented] (SPARK-22147) BlockId.hashCode allocates a
 StringBuilder/String on each call
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Fri, 03 Nov 2017 15:33:07 -0000


    [ https://issues.apache.org/jira/browse/SPARK-22147?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16237775#comment-16237775 ] 

Bryan Cutler commented on SPARK-22147:
--------------------------------------

Sorry, I linked the above PR to this JIRA accidentally

> BlockId.hashCode allocates a StringBuilder/String on each call
> --------------------------------------------------------------
>
>                 Key: SPARK-22147
>                 URL: https://issues.apache.org/jira/browse/SPARK-22147
>             Project: Spark
>          Issue Type: Improvement
>          Components: Block Manager
>    Affects Versions: 2.2.0
>            Reporter: Sergei Lebedev
>            Assignee: Sergei Lebedev
>            Priority: Minor
>             Fix For: 2.3.0
>
>
> The base class {{BlockId}} [defines|https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/storage/BlockId.scala#L44] {{hashCode}} and {{equals}} for all its subclasses in terms of {{name}}. This makes the definitions of different ID types [very concise|https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/storage/BlockId.scala#L52]. The downside, however, is redundant allocations. While I don't think this could be the major issue, it is still a bit disappointing to increase GC pressure on the driver for nothing. For our machine learning workloads, we've seen as much as 10% of all allocations on the driver coming from {{BlockId.hashCode}} calls done for [BlockManagerMasterEndpoint.blockLocations|https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/storage/BlockManagerMasterEndpoint.scala#L54].


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

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