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 momina khan <mo...@gmail.com> on 2008/09/13 14:24:30 UTC

need help with hadoop job stats

hi

if you do a lot of hadoop computation can you tell me how much maximum time
you have encountered with a  hadoop job. can you state the type
of computation (algorithm) and also the cluster setup it was run on, plus
the input data size. we are looking for computations that have taken a large
amount of time.

cheers
momina

Re: need help with hadoop job stats

Posted by Ahmad Humayun <ah...@gmail.com>.
Has anybody run computations over Hadoop, which took days to complete. Just
wondering! I believe that if the scenarios run by Yahoo! (as pointed out by
the web-link) were run on a smaller cluster, this computation might take
days, not hours....is that true?


regards,

On Sat, Sep 13, 2008 at 9:15 PM, Arun C Murthy <ac...@yahoo-inc.com> wrote:

>
> On Sep 13, 2008, at 5:24 AM, momina khan wrote:
>
>  hi
>>
>> if you do a lot of hadoop computation can you tell me how much maximum
>> time
>> you have encountered with a  hadoop job. can you state the type
>> of computation (algorithm) and also the cluster setup it was run on, plus
>> the input data size. we are looking for computations that have taken a
>> large
>> amount of time.
>>
>>
>
> http://developer.yahoo.com/blogs/hadoop/2008/02/yahoo-worlds-largest-production-hadoop.html
>  ?
>
> Arun
>
>  cheers
>> momina
>>
>
>


-- 
Ahmad Humayun
Research Associate
Computer Science Dpt., LUMS
http://suraj.lums.edu.pk/~ahmadh
+92 321 4457315

Re: need help with hadoop job stats

Posted by Arun C Murthy <ac...@yahoo-inc.com>.
On Sep 13, 2008, at 5:24 AM, momina khan wrote:

> hi
>
> if you do a lot of hadoop computation can you tell me how much  
> maximum time
> you have encountered with a  hadoop job. can you state the type
> of computation (algorithm) and also the cluster setup it was run on,  
> plus
> the input data size. we are looking for computations that have taken  
> a large
> amount of time.
>

http://developer.yahoo.com/blogs/hadoop/2008/02/yahoo-worlds-largest-production-hadoop.html 
  ?

Arun

> cheers
> momina