You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by Praveen Sripati <pr...@gmail.com> on 2011/12/07 17:58:57 UTC
Re: Re: how to select the position of a block
Not sure why, but I was able to find ReplicationTargetChooser.java only in
branch-1.0 and not in trunk, 21, 22 and 23.
Regards,
Praveen
2011/12/7 郭冲 <gu...@126.com>
>
>
> isGoodTarget() in ReplicationTargetChooser.java judge if a node is a
> good node, it will consider the disk space , load , and the rack which it
> located has too much nodes or not .Thanks!
>
> At 2011-12-07 13:01:26,"Harsh J" <ha...@cloudera.com> wrote:
> >Moving discussion to hdfs-dev.
> >
> >DataNodes report disk usage (space) and current transfer thread counts
> (load) periodically to the NameNode. NameNode uses this information to make
> a decision while building the pipeline DNs for your client request. I
> believe the class is called something like ReplicationTargetChooser (off
> the top of my mind), and you can take a look at the logic involved there in
> selecting any form of node (isGoodTarget or summat). Then work your way
> downwards to see how the information flows.
> >
> >On 07-Dec-2011, at 8:36 AM, 郭冲 wrote:
> >
> >> Hadoop:The Definitive Guide said that when the client is out of the
> cluster, hadoop will select the storage position of a block randomly,but it
> will not select the datanode which is too busy or too full.
> >>
> >> so i want to kown that how hadoop judge or measure a datanode is full
> or not?
> >
>
Re: Re: how to select the position of a block
Posted by Harsh J <ha...@cloudera.com>.
Yep, its been moved to a different area after pluggable block
placement policies were introduced. I answered with RTC cause OP was
reading the hadoop book, which surrounds what is today 1.0.
2011/12/7 Praveen Sripati <pr...@gmail.com>:
> Not sure why, but I was able to find ReplicationTargetChooser.java only in
> branch-1.0 and not in trunk, 21, 22 and 23.
>
> Regards,
> Praveen
>
> 2011/12/7 郭冲 <gu...@126.com>
>
>>
>>
>> isGoodTarget() in ReplicationTargetChooser.java judge if a node is a
>> good node, it will consider the disk space , load , and the rack which it
>> located has too much nodes or not .Thanks!
>>
>> At 2011-12-07 13:01:26,"Harsh J" <ha...@cloudera.com> wrote:
>> >Moving discussion to hdfs-dev.
>> >
>> >DataNodes report disk usage (space) and current transfer thread counts
>> (load) periodically to the NameNode. NameNode uses this information to make
>> a decision while building the pipeline DNs for your client request. I
>> believe the class is called something like ReplicationTargetChooser (off
>> the top of my mind), and you can take a look at the logic involved there in
>> selecting any form of node (isGoodTarget or summat). Then work your way
>> downwards to see how the information flows.
>> >
>> >On 07-Dec-2011, at 8:36 AM, 郭冲 wrote:
>> >
>> >> Hadoop:The Definitive Guide said that when the client is out of the
>> cluster, hadoop will select the storage position of a block randomly,but it
>> will not select the datanode which is too busy or too full.
>> >>
>> >> so i want to kown that how hadoop judge or measure a datanode is full
>> or not?
>> >
>>
--
Harsh J
RE: Re: how to select the position of a block
Posted by Uma Maheswara Rao G <ma...@huawei.com>.
Praveen,
It has re-factored in trunk.
You can find this in BlockPlacementPolicy ( interface) implementation class. i.e BlockPlacementPolicyDefault.
Regards,
Uma
________________________________________
From: Praveen Sripati [praveensripati@gmail.com]
Sent: Wednesday, December 07, 2011 10:28 PM
To: common-dev@hadoop.apache.org
Subject: Re: Re: how to select the position of a block
Not sure why, but I was able to find ReplicationTargetChooser.java only in
branch-1.0 and not in trunk, 21, 22 and 23.
Regards,
Praveen
2011/12/7 郭冲 <gu...@126.com>
>
>
> isGoodTarget() in ReplicationTargetChooser.java judge if a node is a
> good node, it will consider the disk space , load , and the rack which it
> located has too much nodes or not .Thanks!
>
> At 2011-12-07 13:01:26,"Harsh J" <ha...@cloudera.com> wrote:
> >Moving discussion to hdfs-dev.
> >
> >DataNodes report disk usage (space) and current transfer thread counts
> (load) periodically to the NameNode. NameNode uses this information to make
> a decision while building the pipeline DNs for your client request. I
> believe the class is called something like ReplicationTargetChooser (off
> the top of my mind), and you can take a look at the logic involved there in
> selecting any form of node (isGoodTarget or summat). Then work your way
> downwards to see how the information flows.
> >
> >On 07-Dec-2011, at 8:36 AM, 郭冲 wrote:
> >
> >> Hadoop:The Definitive Guide said that when the client is out of the
> cluster, hadoop will select the storage position of a block randomly,but it
> will not select the datanode which is too busy or too full.
> >>
> >> so i want to kown that how hadoop judge or measure a datanode is full
> or not?
> >
>