Rack Configuration::!

[Sugandha Naolekar <su...@gmail.com>]

Hello!

How to configure the machines in different racks?

I have in all 10 machines.

Now I want the heirarchy as follows::

machine1
machine2
machine3----------these are all DN and TT
machine4

machine5 ----->JT1

machine7
machine8------> JT2

machine10---->NN and Sec.NN


As of now I have 7 machines running a hadoop cluster. which follws below
heirarchy::

machine1
machine2------> these are all DN and TT
machine3
machine4

machine5------>JT

machine6----->Sec.NN

machine7--->NN

Also, if the machines are configured in different racks, what advantage do
we have? Also, give me few problem statements which handles big amount of
data(processing). What Yahoo and Amazon guys have done? What kind of huge
processing of huge data they have handled?


-- 
Regards!
Sugandha

Re: Rack Configuration::!

[Harish Mallipeddi <ha...@gmail.com>]

On Tue, Jun 16, 2009 at 2:45 PM, Sugandha Naolekar
<su...@gmail.com>wrote:

> Hello!
>
> How to configure the machines in different racks?
>
>
> Also, if the machines are configured in different racks, what advantage do
> we have?


See this thread:
http://www.nabble.com/Hadoop-topology.script.file.name-Form-td17683521.html


> Also, give me few problem statements which handles big amount of
> data(processing). What Yahoo and Amazon guys have done? What kind of huge
> processing of huge data they have handled?
>
>
Just google "map reduce applications". Read the original map-reduce paper by
Googlers.

-- 
Harish Mallipeddi
http://blog.poundbang.in