You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@mahout.apache.org by Lukáš Vlček <lu...@gmail.com> on 2009/06/15 22:40:28 UTC

Pregel - large.scale graph computation

Hi,
Does anybody have any details about Pregel? According to the following blog
post the Pregel seems to be Google's way how to handle parallel computation
on top of graph data.
http://googleresearch.blogspot.com/2009/06/large-scale-graph-computing-at-google.html

<http://googleresearch.blogspot.com/2009/06/large-scale-graph-computing-at-google.html>
Regards,
Lukas

Re: Pregel - large.scale graph computation

Posted by Grant Ingersoll <gs...@apache.org>.

Lukas,

This is very cool.  I've long had an interest in graph stuff and not  
just for PageRank, either.  Turns out you can do some fun NLP things  
with graphs (http://www.textgraphs.org/ws09/index.html).  Sounds like  
a nice thing to add to Hadoop or Mahout once more details are  
published.  In the meantime, it's not all that hard to write up some  
basic graph ranking stuff given either HBase or the Matrix stuff in  
Mahout.

-Grant

On Jun 15, 2009, at 4:40 PM, Lukáš Vlček wrote:

> Hi,
> Does anybody have any details about Pregel? According to the  
> following blog
> post the Pregel seems to be Google's way how to handle parallel  
> computation
> on top of graph data.
> http://googleresearch.blogspot.com/2009/06/large-scale-graph-computing-at-google.html
>
> <http://googleresearch.blogspot.com/2009/06/large-scale-graph-computing-at-google.html 
> >
> Regards,
> Lukas