hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Hyunsik Choi <hyunsik.c...@gmail.com>
Subject Re: Above all, I plan to make BSP more complete
Date Sun, 28 Mar 2010 16:45:42 GMT
As you said, the discussion on pregel paper will be very important.
Besides, we could accomplish the survey and the implementation of  BSP
package before the paper is available. It is possible because BSP
package is already known. We can find some literatures on BSP
computation model in Amazon and google scholar.

I'm willing to welcome online and offline discussion. But, I prefer a
mailling list discussion a little more than an offline meeting.

Thank you for your good advice :)
Hyunsik Choi
Database & Information Systems Group, Korea Univ.

On Sun, Mar 28, 2010 at 11:48 PM, Sangwon Seo <smiler.seo@gmail.com> wrote:
> I believe your good intuition.
> As HBase and Hadoop regarded as a valuable project by cloning Google's
> papers,
> our BSP package also have a opportunity to become a outstanding project in a
> similar way.
> Let's discuss pregel paper in detail if the paper is released.
> If possible I want to make meeting in offline as well as online.
> In addition, I think that we first take into account performance issues
> before discussing heterogeneity and fault tolerance (just my opinion).
> Maybe, one of my collegue has scheduled to attend the SIGMOD 2010 as an
> author (regular paper).
> I hope He will give us good information.
> 2010/3/28 Hyunsik Choi <hyunsik.choi@gmail.com>
>> Hi,
>> I would like to discuss the plan that I think.
>> First, I participated in hama in order to contribute graph package. I
>> plan to develop graph package (called Angrapa) for large scale graph
>> data running on shared-nothing architecture. Many of people think
>> about that Angrapa may be appropriate in BSP model, like pregel, and I
>> also agree with that thinking. In addition, the pregel seems to be
>> proven by ACM SIGMOD Conf. 2010.
>> The BSP package is the base of Angrapa. Thus, we first should complete
>> BSP package. Besides, we should be concerned with two design
>> considerations caused by large-scale data and shared-nothing
>> architecture, recently called cloud computing environments. They are
>> heterogeneity and fault tolerance respectively. In addition, it is
>> important to design BSP to be general purpose since it will be used
>> for matrix package and other problems.
>> After pregel paper is published, we should embrace the techniques
>> discussed in the pregel paper.
>> In sum,
>>  * We should complete BSP package, general purpose as possible.
>>  * BSP package has to include considerations of both heterogeneity and
>> fault tolerance.
>> I need advices.
>> Best wishes,
>> --
>> Hyunsik Choi
>> Database & Information Systems Group, Korea Univ.
>> http://diveintodata.org
> --
> Best Regards,
> Sangwon Seo.
> --------------------------------------------------------
>  Sangwon Seo,
> Computer Architecture Lab,
> http://smiler-note.blogspot.com

View raw message