Re: Config. Name Node and Job Tracker in the same machine

[Ravi Phulari <rp...@yahoo-inc.com>]

If your cluster is not big enough (thousands of nodes) then there should be no problem in running NN & JT on same machine.
I usually run my smaller clusters of size 10-20 nodes with NN & JT on same machine.

Usually bigger clusters have millions of HDFS files and it takes lots of NN memory that's why it is suggested to use different machines for NN & JT.

Thanks,
Ravi

On 3/29/10 1:01 PM, "Edson Ramiro" <er...@gmail.com> wrote:

Hi all,

Is there any problem in configure NameNode and JobTracker in the same
machine?

The documentation[1] Says that:

"Typically one machine in the cluster is designated as the NameNode and
another
machine the as JobTracker..."

But it doesn't say that we need to have one machine to each service.

Thanks in Advance.

[1] http://hadoop.apache.org/common/docs/current/cluster_setup.html

Edson Ramiro

Re: Config. Name Node and Job Tracker in the same machine

[Edson Ramiro <er...@gmail.com>]

Hi Ravi,

I think Its not explicit in the documentation.

I've 26 nodes and 1,9 Tb of disc and I'll try to run it on same machine to
test.

Thanks for reply

Edson Ramiro


On 29 March 2010 17:38, Ravi Phulari <rp...@yahoo-inc.com> wrote:

> If your cluster is not big enough (thousands of nodes) then there should be
> no problem in running NN & JT on same machine.
> I usually run my smaller clusters of size 10-20 nodes with NN & JT on same
> machine.
>
> Usually bigger clusters have millions of HDFS files and it takes lots of NN
> memory that's why it is suggested to use different machines for NN & JT.
>
> Thanks,
> Ravi
>
> On 3/29/10 1:01 PM, "Edson Ramiro" <er...@gmail.com> wrote:
>
> Hi all,
>
> Is there any problem in configure NameNode and JobTracker in the same
> machine?
>
> The documentation[1] Says that:
>
> "Typically one machine in the cluster is designated as the NameNode and
> another
> machine the as JobTracker..."
>
> But it doesn't say that we need to have one machine to each service.
>
> Thanks in Advance.
>
> [1] http://hadoop.apache.org/common/docs/current/cluster_setup.html
>
> Edson Ramiro
>
>
>