Return-Path: X-Original-To: apmail-hadoop-hdfs-user-archive@minotaur.apache.org Delivered-To: apmail-hadoop-hdfs-user-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 4FF7818686 for ; Sun, 23 Aug 2015 20:30:08 +0000 (UTC) Received: (qmail 70222 invoked by uid 500); 23 Aug 2015 20:30:03 -0000 Delivered-To: apmail-hadoop-hdfs-user-archive@hadoop.apache.org Received: (qmail 70078 invoked by uid 500); 23 Aug 2015 20:30:02 -0000 Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@hadoop.apache.org Delivered-To: mailing list user@hadoop.apache.org Received: (qmail 70068 invoked by uid 99); 23 Aug 2015 20:30:02 -0000 Received: from Unknown (HELO spamd3-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Sun, 23 Aug 2015 20:30:02 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd3-us-west.apache.org (ASF Mail Server at spamd3-us-west.apache.org) with ESMTP id 32285182345 for ; Sun, 23 Aug 2015 20:30:02 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd3-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: 3.879 X-Spam-Level: *** X-Spam-Status: No, score=3.879 tagged_above=-999 required=6.31 tests=[DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_REPLY=1, HTML_MESSAGE=3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=disabled Authentication-Results: spamd3-us-west.apache.org (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com Received: from mx1-us-east.apache.org ([10.40.0.8]) by localhost (spamd3-us-west.apache.org [10.40.0.10]) (amavisd-new, port 10024) with ESMTP id ojQUMppPbrNh for ; Sun, 23 Aug 2015 20:30:00 +0000 (UTC) Received: from mail-oi0-f54.google.com (mail-oi0-f54.google.com [209.85.218.54]) by mx1-us-east.apache.org (ASF Mail Server at mx1-us-east.apache.org) with ESMTPS id 9BD7C42996 for ; Sun, 23 Aug 2015 20:29:59 +0000 (UTC) Received: by oiey141 with SMTP id y141so68986633oie.1 for ; Sun, 23 Aug 2015 13:29:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=KFEwP09HhDRniAHwzagMKGrQE/yn8SCM0j6YAxDUV1A=; b=XH3fM+dJOhqw56hZc4GUiDNQ5mIpL8jxGuNs5lMHIa3b8RZYqq8vcW+5v9o53gAz5B fMN5GDs5hkMS9nUizQ64GivAfpOY1r+ClJIODGRXDBPag/1OX2U4CPXiUnXFAP1XjlJN r1H5qK2pgktx4YVcqcIAdi/P2QH3eISm/nEbkGYbjdOEXw1DGTNLDYU93QYBMz0yKNiz +YubI305iEKGrfPlw5Nlc5aHegy8/lBCWr0JZwWfTIw8sKoDWZ8afY9vNBdIZ801Tfaf 3dtMlpeYVPVvWJAK6V7K1nXvE9gebRDgAIv0qJ8sPigfjBA3G0xjJEi5vDhrcbyxbk/Z NCWw== MIME-Version: 1.0 X-Received: by 10.202.215.8 with SMTP id o8mr17352254oig.47.1440361793483; Sun, 23 Aug 2015 13:29:53 -0700 (PDT) Received: by 10.76.28.36 with HTTP; Sun, 23 Aug 2015 13:29:53 -0700 (PDT) In-Reply-To: References: Date: Mon, 24 Aug 2015 01:59:53 +0530 Message-ID: Subject: Re: yarn.nodemanager.resource.cpu-vcores vs yarn.scheduler.maximum-allocation-vcores From: Varun Saxena To: user@hadoop.apache.org Content-Type: multipart/alternative; boundary=001a113d38020c7f66051e005d90 --001a113d38020c7f66051e005d90 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable So how does hadoop get this property if it is per node? Does it get the minimum of all nodes? --> No its not minimum of all nodes. Each nodemanager reads this configuration from its respective configuration file(yarn-site.xml). Nodemanager is like an agent which manages the lifecycle of containers and installed on each node where you want to run containers. It communicates with resource manager and that is how resource manager comes to know about capability of each node. At the time of registration with RM, Nodemanager tells about that node's capability to RM(for scheduling) by reading above 2 configuration items(one for memory and one for vcores). By capability of node I meant you may have some nodes which has 8 cores and some which have 16 cores, for instance. Some may have 16 GB memory and some 24 GB. So above 2 configurations can be configured accordingly because till Hadoop 2.7 we were not getting a node's hardware capability from operating system. This will be automatically read from OS(Linux/Windows), if configured to do so, from 2.8 onwards. This is a nodemanager configuration and is not required to be configured at the client side while submitting the job. Regards, Varun Saxena On Mon, Aug 24, 2015 at 1:26 AM, Varun Saxena wrote: > This configuration is read and used by NodeManager, on whichever node its > running. > If it is not configured, default value will be taken. > > Regards, > Varun Saxena. > > On Mon, Aug 24, 2015 at 1:21 AM, Pedro Magalhaes > wrote: > >> Thanks Varun! Like we say in Brazil. "U are the guy!" (Voc=C3=AA =C3=A9= o cara!) >> >> I have another question. You said that: >> "yarn.nodemanager.resource.cpu-vcores on the other hand will have to be >> configured as per resource capability of that particular node. " >> >> I get the configuration from my job and printed it: >> yarn.nodemanager.resource.cpu-vcores 8 >> yarn.nodemanager.resource.memory-mb 8192 >> >> So how does hadoop get this property if it is per node? Does it get the >> minimum of all nodes? Thanks again! >> >> >> >> On Sun, Aug 23, 2015 at 4:40 PM, Varun Saxena >> wrote: >> >>> The fix would be released in next version(2.8.0). >>> I had checked the code to find out the default value and then found it >>> fixed in documentation(configuration list). >>> >>> As this is an unreleased version, a URL link (of the form >>> https://hadoop.apache.org/docs/r2.7.1/hadoop-yarn/hadoop-yarn-common/ya= rn-default.xml) >>> may not be available AFAIK, >>> However, this XML(yarn-default.xml) can be checked online in git >>> repository. >>> >>> Associated JIRA which fixes this is >>> https://issues.apache.org/jira/browse/YARN-3823 >>> >>> Regards, >>> Varun Saxena. >>> >>> On Mon, Aug 24, 2015 at 12:53 AM, Pedro Magalhaes >>> wrote: >>> >>>> Thanks Varun! >>>> Could plz send me the link with the fixed? >>>> >>>> On Sun, Aug 23, 2015 at 2:20 PM, Varun Saxena >>>> wrote: >>>> >>>>> Hi Pedro, >>>>> >>>>> Real default value of yarn.scheduler.maximum-allocation-vcores is 4. >>>>> The value of 32 is actually a documentation issue and has been fixed >>>>> recently. >>>>> >>>>> Regards, >>>>> Varun Saxena. >>>>> >>>>> >>>>> On Sun, Aug 23, 2015 at 10:39 PM, Pedro Magalhaes >>>> > wrote: >>>>> >>>>>> Varun, >>>>>> Thanks for the reply. I undestand the arn.scheduler.maximum- >>>>>> allocation-vcores parameter. I just asking why the default parameter >>>>>> is yarn.scheduler.maximum-allocation-vcores=3D32. And >>>>>> yarn.nodemanager.resource.cpu-vcores=3D8. >>>>>> >>>>>> In my opinion, if the yarn.scheduler.maximun-allocation-vcore is 32 >>>>>> tby default the yarn.nodemanager.resource.cpu-vcores would be equal= or >>>>>> greater than 32, by default. >>>>>> Is this make sense? >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> On Sun, Aug 23, 2015 at 2:00 PM, Varun Saxena < >>>>>> vsaxena.varun@gmail.com> wrote: >>>>>> >>>>>>> Hi Pedro, >>>>>>> >>>>>>> Actual allocation would depend on the total resource capability >>>>>>> advertised by NM while registering with RM. >>>>>>> >>>>>>> yarn.scheduler.maximum-allocation-vcores merely puts an upper cap o= n number of vcores which can be allocated by RM i.e. any Resource request/a= sk from AM which asks for vcores > 32(default value) for a container, will = be normalized back to 32. >>>>>>> >>>>>>> If there is no such node available, this allocation will not be ful= filled. >>>>>>> >>>>>>> yarn.scheduler.maximum-allocation-vcores will be configured in >>>>>>> resource manager and hence will be common for a cluster which can p= ossibly >>>>>>> have multiple nodes with heterogeneous resource capabilities >>>>>>> >>>>>>> yarn.nodemanager.resource.cpu-vcores on the other hand will have to >>>>>>> be configured as per resource capability of that particular node. >>>>>>> >>>>>>> Recently there has been work done to automatically get memory and >>>>>>> CPU information from underlying OS(supported OS being Linux and Win= dows) if >>>>>>> configured to do so. This change would be available in 2.8 >>>>>>> I hope this answers your question. >>>>>>> >>>>>>> Regards, >>>>>>> Varun Saxena. >>>>>>> >>>>>>> On Sun, Aug 23, 2015 at 9:40 PM, Pedro Magalhaes < >>>>>>> pedrorjbr@gmail.com> wrote: >>>>>>> >>>>>>>> I was looking at default parameters for: >>>>>>>> >>>>>>>> yarn.nodemanager.resource.cpu-vcores =3D 8 >>>>>>>> yarn.scheduler.maximum-allocation-vcores =3D 32 >>>>>>>> >>>>>>>> For me this two parameters as default doesnt make any sense. >>>>>>>> >>>>>>>> The first one say "the number of CPU cores that can be allocated >>>>>>>> for containers." (I imagine that is vcore) The seconds says: "The = maximum >>>>>>>> allocation for every container request at the RM". In my opinion, = the >>>>>>>> second one must be equal or less than the first one. >>>>>>>> >>>>>>>> How can allocate 32 vcores for a container if i have only 8 cores >>>>>>>> available per container? >>>>>>>> >>>>>>> >>>>>>> >>>>>> >>>>> >>>> >>> >> > --001a113d38020c7f66051e005d90 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
So how does hadoop get this property if it is per node? Does it get= the minimum of all nodes?

--> No its not minimum of all nodes.= Each nodemanager reads this configuration from its respective configuratio= n file(yarn-site.xml).=C2=A0
Nodemanager is like an agent which man= ages the lifecycle of containers and installed on each node where you want = to run containers.=C2=A0
It communicates with resource manager = and that is how resource manager comes to know about capability of each nod= e. At the time of registration with RM, Nodemanager tells about that node&#= 39;s capability to RM(for scheduling) by reading above 2 configuration item= s(one for memory and one for vcores).

By capability= of node I meant you may have some nodes which has 8 cores and some which h= ave 16 cores, for instance. Some may have 16 GB memory and some 24 GB.
So above 2 configurations can be configured accordingly because till Ha= doop 2.7 we were not getting a node's hardware capability from operatin= g system. This will be automatically read from OS(Linux/Windows), if config= ured to do so, from 2.8 onwards.

This is a nodemanag= er configuration and is not required to be configured at the client side wh= ile submitting the job.

Regards,
V= arun Saxena


On Mon, Aug 24, 2015 at 1:26 AM, Varun Saxena <vsaxena.varun@gmail.com> wrote:
This configuration is read and used by NodeManage= r, on whichever node its running.
If it is not configured, default valu= e will be taken.

Regards,
Varun Saxena.<= /div>

On Mon, Aug 24, 2015 at 1:21 AM, Pedro Ma= galhaes <pedrorjbr@gmail.com> wrote:
Thanks Varun! Like we say in Brazil. =C2=A0&= quot;U are the guy!" (Voc=C3=AA =C3=A9 o cara!)

I have another = question. You said that:
"yarn.nodemanager.resour= ce.cpu-vcores on the ot= her hand will have to be configured as per resource capability of that part= icular node.=C2=A0"

I get the configuration from my job and printed it:
yarn.nodemanager.resource.cpu-vcores 8=20 yarn.nodemanager.resource.memory-mb 8192

So how does hadoop get this property if it is per node? Does it g= et the minimum of all nodes? Thanks again!



On Sun, Aug 23, 2015 at 4:40 PM, Varun Saxena <vsaxena.varun@gmail.com> wrote:
The fix would be released in next version(2.8.0).= =C2=A0
I had checked the code to find out the default value and then fo= und it fixed in documentation(configuration list).=C2=A0

=
As this is an unreleased version, a URL link (of the form https://hadoop.apache.org/docs/r2.7.1/ha= doop-yarn/hadoop-yarn-common/yarn-default.xml) may not be available AFA= IK,
However, this XML(yarn-default.xml) can be checked online in = git repository.

Associated JIRA which fixes this i= s=C2=A0https://issues.apache.org/jira/browse/YARN-3823
<= br>
Regards,
Varun Saxena.

On Mon, Aug 24, 2015 at = 12:53 AM, Pedro Magalhaes <pedrorjbr@gmail.com> wrote:
=
Thanks Varun!
Could plz= send me the link with the fixed?

On Sun, Aug 23, 2015 at 2:20 PM, Varu= n Saxena <vsaxena.varun@gmail.com> wrote:
Hi Pedro,=

Real = default value of yarn.scheduler.maximum-allocation-vcores is 4.
The value of 32 is actually a documentation issue and = has been fixed recently.

Regards,
Varun Saxena.


On Sun, Aug 23, 2015 at 10:39 PM, Pedro M= agalhaes <pedrorjbr@gmail.com> wrote:
Varun,
Thanks for the reply. I undestand= the=C2=A0arn.scheduler.m= aximum-allocation-= vcores=C2=A0parameter. I just asking why the default parameter is yarn.scheduler.maximum-allocation-vcores=3D32= .=C2=A0And=C2=A0ya= rn.nodemanager.resource.cpu-vcores=3D8.

In my opinion, if the= yarn.scheduler.maximun-allocation-vcore is 32 tby default the yarn.nodeman= ager.resource.cpu-vcores =C2=A0would be equal or greater than 32, by defaul= t.
Is this make sense?



On Sun, Aug 23, 2015 at 2:00 PM, Varun Saxena <= vsaxena.varun@= gmail.com> wrote:
Hi Pedro,

Actual allocation would depend on the= total resource capability advertised by NM while registering with RM.
yarn.scheduler.maximum-allocation-vco=
res merely puts an upper cap on number of vcores which can be allocated by =
RM i.e. any Resource request/ask from AM which asks for vcores > 32(defa=
ult value) for a container, will be normalized back to 32.
If there is no such node available, this alloca= tion will not be fulfilled.
yarn.scheduler.maximum-a= llocation-vcores will be configured in resource manager and hence will be c= ommon for a cluster which can possibly have multiple nodes with heterogeneo= us resource capabilities

yarn.nodemanager.resource= .cpu-vcores on the other hand will have to be configured as per resource ca= pability of that particular node.=C2=A0

Recently t= here has been work done to automatically get memory and CPU information fro= m underlying OS(supported OS being Linux and Windows) if configured to do s= o. This change would be available in 2.8
I hope this answers your= question.

Regards,
Varun Saxena.
<= /div>

On= Sun, Aug 23, 2015 at 9:40 PM, Pedro Magalhaes <pedrorjbr@gmail.com&= gt; wrote:

I was looking at default parameters for:

yarn.nodeman=
ager.resource.cpu-vcores =3D 8
yarn.scheduler.maximum-allocation-vcores =3D 32

For me this two parameters as default doesnt mak= e any sense.

The first one say "the number of CPU cor= es that can be allocated for containers." (I imagine that is vcore) Th= e seconds says: "The maximum allocation for every container request at= the RM". In my opinion, the second one must be equal or less than the= first one.

How can allocate 32 vcores for a container if = i have only 8 cores available per container?









--001a113d38020c7f66051e005d90--