You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hadoop.apache.org by John Lilley <jo...@redpoint.net> on 2013/10/06 00:12:01 UTC
MapReduce task-worker assignment
Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign "one HDFS block per mapper task", but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
Thanks,
John Lilley
Chief Architect, RedPoint Global Inc.
1515 Walnut Street | Suite 200 | Boulder, CO 80302
T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
Skype: jlilley.redpoint | john.lilley@redpoint.net<ma...@redpoint.net> | www.redpoint.net<http://www.redpoint.net/>
RE: MapReduce task-worker assignment
Posted by John Lilley <jo...@redpoint.net>.
Thanks!
From: Arun C Murthy [mailto:acm@hortonworks.com]
Sent: Monday, October 07, 2013 6:07 PM
To: user@hadoop.apache.org
Subject: Re: MapReduce task-worker assignment
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard... *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net>> wrote:
Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign "one HDFS block per mapper task", but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
Thanks,
John Lilley
Chief Architect, RedPoint Global Inc.
1515 Walnut Street | Suite 200 | Boulder, CO 80302
T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
Skype: jlilley.redpoint | john.lilley@redpoint.net<ma...@redpoint.net> | www.redpoint.net<http://www.redpoint.net/>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You.
RE: MapReduce task-worker assignment
Posted by John Lilley <jo...@redpoint.net>.
Thanks!
From: Arun C Murthy [mailto:acm@hortonworks.com]
Sent: Monday, October 07, 2013 6:07 PM
To: user@hadoop.apache.org
Subject: Re: MapReduce task-worker assignment
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard... *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net>> wrote:
Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign "one HDFS block per mapper task", but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
Thanks,
John Lilley
Chief Architect, RedPoint Global Inc.
1515 Walnut Street | Suite 200 | Boulder, CO 80302
T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
Skype: jlilley.redpoint | john.lilley@redpoint.net<ma...@redpoint.net> | www.redpoint.net<http://www.redpoint.net/>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You.
RE: MapReduce task-worker assignment
Posted by John Lilley <jo...@redpoint.net>.
Thanks!
From: Arun C Murthy [mailto:acm@hortonworks.com]
Sent: Monday, October 07, 2013 6:07 PM
To: user@hadoop.apache.org
Subject: Re: MapReduce task-worker assignment
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard... *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net>> wrote:
Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign "one HDFS block per mapper task", but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
Thanks,
John Lilley
Chief Architect, RedPoint Global Inc.
1515 Walnut Street | Suite 200 | Boulder, CO 80302
T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
Skype: jlilley.redpoint | john.lilley@redpoint.net<ma...@redpoint.net> | www.redpoint.net<http://www.redpoint.net/>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You.
RE: MapReduce task-worker assignment
Posted by John Lilley <jo...@redpoint.net>.
Thanks!
From: Arun C Murthy [mailto:acm@hortonworks.com]
Sent: Monday, October 07, 2013 6:07 PM
To: user@hadoop.apache.org
Subject: Re: MapReduce task-worker assignment
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard... *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net>> wrote:
Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign "one HDFS block per mapper task", but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
Thanks,
John Lilley
Chief Architect, RedPoint Global Inc.
1515 Walnut Street | Suite 200 | Boulder, CO 80302
T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
Skype: jlilley.redpoint | john.lilley@redpoint.net<ma...@redpoint.net> | www.redpoint.net<http://www.redpoint.net/>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You.
Re: MapReduce task-worker assignment
Posted by Arun C Murthy <ac...@hortonworks.com>.
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard… *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net> wrote:
> Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign “one HDFS block per mapper task”, but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
> Thanks,
> John Lilley
> Chief Architect, RedPoint Global Inc.
> 1515 Walnut Street | Suite 200 | Boulder, CO 80302
> T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
> Skype: jlilley.redpoint | john.lilley@redpoint.net | www.redpoint.net
>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.
Re: MapReduce task-worker assignment
Posted by Arun C Murthy <ac...@hortonworks.com>.
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard… *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net> wrote:
> Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign “one HDFS block per mapper task”, but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
> Thanks,
> John Lilley
> Chief Architect, RedPoint Global Inc.
> 1515 Walnut Street | Suite 200 | Boulder, CO 80302
> T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
> Skype: jlilley.redpoint | john.lilley@redpoint.net | www.redpoint.net
>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.
Re: MapReduce task-worker assignment
Posted by Arun C Murthy <ac...@hortonworks.com>.
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard… *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net> wrote:
> Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign “one HDFS block per mapper task”, but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
> Thanks,
> John Lilley
> Chief Architect, RedPoint Global Inc.
> 1515 Walnut Street | Suite 200 | Boulder, CO 80302
> T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
> Skype: jlilley.redpoint | john.lilley@redpoint.net | www.redpoint.net
>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.
Re: MapReduce task-worker assignment
Posted by Arun C Murthy <ac...@hortonworks.com>.
Short version: MR provides all the info it can to (about all it's tasks locations) and the YARN scheduler deals with providing good locality with even assignment.
I don't have a handy link to a doc, but here is an ancient version:
http://developer.yahoo.com/blogs/hadoop/next-generation-apache-hadoop-mapreduce-scheduler-4141.html
You might find a more recent version of one of my talks if you try really hard… *smile*
Arun
On Oct 5, 2013, at 3:12 PM, John Lilley <jo...@redpoint.net> wrote:
> Is there a description of how MapReduce under Hadoop 2.0 assigns mapper tasks to preferred nodes? I think that someone on the list mentioned previously that it attempted to assign “one HDFS block per mapper task”, but given that there can be multiple block instances per data split, how does MapReduce try to obtain an even task assignment while optimizing data locality?
> Thanks,
> John Lilley
> Chief Architect, RedPoint Global Inc.
> 1515 Walnut Street | Suite 200 | Boulder, CO 80302
> T: +1 303 541 1516 | M: +1 720 938 5761 | F: +1 781-705-2077
> Skype: jlilley.redpoint | john.lilley@redpoint.net | www.redpoint.net
>
--
Arun C. Murthy
Hortonworks Inc.
http://hortonworks.com/
--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.