Just FYI, We've already made a prototype of Hamburg based on BSP model.

I guess we could also improve the performance of matrix operations
using BSP computing model.

---------- Forwarded message ----------
From: Edward J. Yoon <[email protected]>
Date: Fri, Sep 4, 2009 at 9:57 AM
Subject: Re: Pregel
To: [email protected]


We've already made a prototype of Hamburg based on multi thread. It's
a BSP based graph computing framework, not a M/R based application.

Please Join to ... http://groups.google.com/group/hamburg-dev

On Fri, Sep 4, 2009 at 8:45 AM, Mark Kerzner<[email protected]> wrote:
> But Ted,
> I am interested specifically in Pregel kind of system, for distributed graph
> operations, and Mahout is for distributed learning. Here what I would
> ideally like to do:
>
>
>   - Somebody must have info on Pregel - it's out, has been presented, and
>   the information is public. Anybody has been there and can at least re-tell?
>   The subscriptions are expensive, but I could buy one copy of the article, if
>   it is available. I will also write directly to the authors;
>   - Study that and discuss relevant information and architecture;
>   - Do the first implementation.
>
> I personally don't like the name Hamburg, but I could live with that.
>
> Mark
>
> On Thu, Sep 3, 2009 at 6:36 PM, Ted Dunning <[email protected]> wrote:
>
>> Hamburg has been excessively "stable" for some time.  If you want to do
>> something, I would recommend contributing to Mahout.
>>
>> On Thu, Sep 3, 2009 at 3:51 PM, Ashutosh Chauhan <
>> [email protected]
>> > wrote:
>>
>> > Hamburg is here: http://wiki.apache.org/hadoop/Hamburg
>> >
>> >
>>
>



--
Best Regards, Edward J. Yoon @ NHN, corp.
[email protected]
http://blog.udanax.org



-- 
Best Regards, Edward J. Yoon @ NHN, corp.
[email protected]
http://blog.udanax.org

Reply via email to