hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Neal Clark <ncl...@uvic.ca>
Subject Re: GSoC 2011 mentoring
Date Wed, 16 Mar 2011 12:01:19 GMT
There are a number of shortest path algorithms. A good place to start
whenever looking for parallel BSP algorithms is the old theory work on PRAM
machines.

http://www.google.ca/search?sourceid=chrome&ie=UTF-8&q=pram+shortest+path+algorithm

There are a few papers listed in the above search which would likely be an
excellent starting point for a GSOC student.

Thanks,

Neal.

On Wed, Mar 16, 2011 at 6:53 AM, Edward J. Yoon <edwardyoon@apache.org>wrote:

> > So to conclude, I think it's a good idea to mention Bellman-Ford and give
> > some resources about it without define it as mandatory (it's good to give
> > students some freedom - it's motivating), then it's the role of the
> mentor
> > to guide the student to the right implementation and how the algorithm
> can
> > be applied to BSP.
>
> +1, I agree with you.
>
> Thanks for your comment and advice!
>
> On Wed, Mar 16, 2011 at 7:08 PM, Alois Cochard <alois.cochard@gmail.com>
> wrote:
> > Since Bellman-Ford seems to be the only distributed shortest path finding
> > algorithm to exist actually (is it correct ?), would be nice to point
> > students to it.
> >
> > But I think it's must not be a 'mandatory point' if a student is capable
> to
> > come with his own distributed algorithm - we don't know, perhaps
> > a genius student will come with this - we can accept it.
> >
> > For the majority of students, telling them to focus on Bellman-Ford is a
> > good idea, making them able to think about implementation and
> understanding
> > Hama/BSP instead of breaking their heads by trying to find an original
> > algorithm (or trying to make distributed a 'non-distributable' one) and
> > loosing lot of time.
> >
> > I found a nice publication about history of Bellman-Ford / Distributed
> > [Asynchronous] Bellman-Ford:
> > http://www.walden-family.com/public/bf-history.pdf
> > (fun to see that this algorithm is used by ARPANET for routing -didn't
> know
> > that- and how the algorithm evolved)
> >
> > So to conclude, I think it's a good idea to mention Bellman-Ford and give
> > some resources about it without define it as mandatory (it's good to give
> > students some freedom - it's motivating), then it's the role of the
> mentor
> > to guide the student to the right implementation and how the algorithm
> can
> > be applied to BSP.
> >
> > I'm more used to Dijkstra/A* than Bellman-Ford and still a newbee with
> Hama
> > but if you need help for mentorship (or as a backup mentor) don't
> hesitate
> > to ask, I have a strong interest for this kind of algorithms/problems and
> > open to spend some of my free time on this.
> >
> > Cheers,
> >
> >
> > On 15 March 2011 15:22, Edward J. Yoon <edwardyoon@apache.org> wrote:
> >
> >> Thanks.
> >>
> >> BTW, What do you think about the task topic? Students have to design
> >> the scenario from A to Z.
> >>
> >> The task description should be detailed and informative? If so, it
> >> seems better to focus on bellman-ford algorithm.
> >>
> >> On Tue, Mar 15, 2011 at 9:13 PM, Alois Cochard <alois.cochard@gmail.com
> >
> >> wrote:
> >> > Thanks a lot Eddie, everything is ok now :)
> >> >
> >> > The shortest path finding algorithm is a really a great project, if
> was
> >> > still a student I would loved working on this !
> >> >
> >> > I hope that a passionate student will find interest in it !
> >> >
> >> > Good luck !
> >> >
> >> > --
> >> > *Alois Cochard*
> >> > http://aloiscochard.blogspot.com
> >> > http://twitter.com/aloiscochard
> >> >
> >> > On 15 March 2011 11:38, Edward J. Yoon <edwardyoon@apache.org> wrote:
> >> >
> >> >> > Even if not qualified for mentor-ship I wanted to take a look
at
> it,
> >> but
> >> >> the
> >> >> > link don't work :(
> >> >>
> >> >> Sounds great!
> >> >>
> >> >> > "The selected filter with id '12315366' does not exist."
> >> >> >
> >> >> > Perhaps I haven't sufficient right in JIRA to see it ?
> >> >>
> >> >> Oh, .. I've just changed permission to everyone, Pls check again.
> >> >>
> >> >> Thanks.
> >> >>
> >> >> On Tue, Mar 15, 2011 at 7:30 PM, Alois Cochard <
> alois.cochard@gmail.com
> >> >
> >> >> wrote:
> >> >> > Hello,
> >> >> >
> >> >> > Even if not qualified for mentor-ship I wanted to take a look
at
> it,
> >> but
> >> >> the
> >> >> > link don't work :(
> >> >> >
> >> >> > "The selected filter with id '12315366' does not exist."
> >> >> >
> >> >> > Perhaps I haven't sufficient right in JIRA to see it ?
> >> >> >
> >> >> > Thanks,
> >> >> >
> >> >> > Alois Cochard
> >> >> > http://aloiscochard.blogspot.com
> >> >> > http://twitter.com/aloiscochard
> >> >> >
> >> >> > On 11 March 2011 02:06, Edward J. Yoon <edwardyoon@apache.org>
> wrote:
> >> >> >
> >> >> >> Here's our gsoc tasks.
> >> >> >>
> >> >> >> http://s.apache.org/Hama-GSoC
> >> >> >>
> >> >> >> Who can participate in these issues as mentor?
> >> >> >>
> >> >> >> I can help for all issues as a mentor or a co-mentor.
> >> >> >>
> >> >> >> Thanks.
> >> >> >> --
> >> >> >> Best Regards, Edward J. Yoon
> >> >> >> http://blog.udanax.org
> >> >> >> http://twitter.com/eddieyoon
> >> >> >>
> >> >> >
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> Best Regards, Edward J. Yoon
> >> >> http://blog.udanax.org
> >> >> http://twitter.com/eddieyoon
> >> >>
> >> >
> >>
> >>
> >>
> >> --
> >> Best Regards, Edward J. Yoon
> >> http://blog.udanax.org
> >> http://twitter.com/eddieyoon
> >>
> >
> >
> >
> > --
> > *Alois Cochard*
> > http://aloiscochard.blogspot.com
> > http://twitter.com/aloiscochard
> >
>
>
>
> --
> Best Regards, Edward J. Yoon
> http://blog.udanax.org
> http://twitter.com/eddieyoon
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message