Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Thu, 26 Oct 2017 20:42:00 +0000 (UTC)
From: "Misha Dmitriev (JIRA)" <jira@apache.org>
To: yarn-issues@hadoop.apache.org
Message-ID: <JIRA.13108983.1507829469000.72736.1509050520577@Atlassian.JIRA>
In-Reply-To: <JIRA.13108983.1507829469000@Atlassian.JIRA>
References: <JIRA.13108983.1507829469000@Atlassian.JIRA> <JIRA.13108983.1507829469956@jira-lw-us.apache.org>
Subject: [jira] [Commented] (YARN-7320) Duplicate LiteralByteStrings in
 SystemCredentialsForAppsProto.credentialsForApp_
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
archived-at: Thu, 26 Oct 2017 20:42:06 -0000


    [ https://issues.apache.org/jira/browse/YARN-7320?page=3Dcom.atlassian.=
jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D16221=
206#comment-16221206 ]=20

Misha Dmitriev commented on YARN-7320:
--------------------------------------

[~rkanter] [~wangda] looks like in ~24 hrs Jenkins still hasn't processed m=
y patch. Could you please check what's going on?

> Duplicate LiteralByteStrings in SystemCredentialsForAppsProto.credentials=
ForApp_
> -------------------------------------------------------------------------=
-------
>
>                 Key: YARN-7320
>                 URL: https://issues.apache.org/jira/browse/YARN-7320
>             Project: Hadoop YARN
>          Issue Type: Improvement
>            Reporter: Misha Dmitriev
>            Assignee: Misha Dmitriev
>             Fix For: 3.0.0
>
>         Attachments: YARN-7320.01.addendum.patch, YARN-7320.01.patch, YAR=
N-7320.02.patch
>
>
> Using jxray (www.jxray.com) I've analyzed several heap dumps from YARN Re=
source Manager running in a big cluster. The tool uncovered several sources=
 of memory waste. One problem, which results in wasting more than a quarter=
 of all memory, is a large number of duplicate {{LiteralByteString}} object=
s coming from the following reference chain:
> {code}
> 1,011,810K (26.9%): byte[]: 5416705 / 100% dup arrays (22108 unique)
> =E2=86=96com.google.protobuf.LiteralByteString.bytes
> =E2=86=96org.apache.hadoop.yarn.proto.YarnServerCommonServiceProtos$.cred=
entialsForApp_
> =E2=86=96{j.u.ArrayList}
> =E2=86=96j.u.Collections$UnmodifiableRandomAccessList.c
> =E2=86=96org.apache.hadoop.yarn.proto.YarnServerCommonServiceProtos$NodeH=
eartbeatResponseProto.systemCredentialsForApps_
> =E2=86=96org.apache.hadoop.yarn.server.api.protocolrecords.impl.pb.NodeHe=
artbeatResponsePBImpl.proto
> =E2=86=96org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeImpl.=
latestNodeHeartBeatResponse
> =E2=86=96org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FSS=
chedulerNode.rmNode
> ...
> {code}
> That is, collectively reference chains that look as above hold in memory =
5.4 million {{LiteralByteString}} objects, but only ~22 thousand of these o=
bjects are unique. Deduplicating these objects, e.g. using a Google Object =
Interner instance, would save ~1GB of memory.
> It looks like the main place where the above {{LiteralByteString}}s are c=
reated and attached to the {{SystemCredentialsForAppsProto}} objects is in =
{{NodeHeartbeatResponsePBImpl.java}}, method {{addSystemCredentialsToProto(=
)}}. Probably adding a call to an interner there will fix the problem. wi=
=20


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

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