Boost logo

Boost :

Subject: Re: [boost] GSoC 2015 prospective student
From: Tanuj kr Aasawat (tanujkraasawat_at_[hidden])
Date: 2015-03-19 09:43:21


Hi Mikael,
              Actually my class tests are going on, that's why I'm a bit
short of time and not able to see the implementations. But I have some
ideas like, I implemented Label propagation algorithm(parallel cpu version)
when I was at University of British Columbia, Canada as Mitacs Globalink
Scholar. Practical application of label propagation algorithm is "People
You May Know Service" of different social networking sites. Another
algorithm on which I co-worked there at UBC, was computing clustering
coefficient parallelly using GPUs(with Nvidia CUDA using C). Our work
outperformed any other by 7x speedup and it was accepted for poster
presentation in prestigious Super Computing Conference, SC'14, New Orleans,
LA. Clustering coefficient represents how tightly vertices are bounded in a
graph.
As I am going for further studies in the field of high performance
computing(parallel and distributed), I am willing to work on graph
algorithms. It would be of great help if you could suggest me some projects
that I should be working on as GSoC under your mentorship. In the mean
time, I would also be going through the implementations and other ideas
that could be implemented.
My final exams will be over by 18th of May, 2015 and I would be going for
my Masters in September, so I would be available full time.

My programming skills in C/C++:
 C, C++ >> (since 1st year) >> Also used them during my research internship
at IIT-KGP. Apart from that, recently, my team was the finalist in "Overnite
ACM ICPC Multi Provincial Programming Contest" which was held at IIT-KGP.

Anticipating positive reply from your side.

Thanks and Regards,
Tanuj Kr Aasawat
Deptt. of I.T, Jadavpur University, India
Mitacs Globalink Research Intern,
University of British Columbia, Vancouver, Canada

On Wed, Mar 18, 2015 at 2:12 AM, Mikael Persson <mikael.s.persson_at_[hidden]>
wrote:

> Hi Tanuj,
>
> Boost already has a comprehensive graph library, in Boost.Graph:
> http://www.boost.org/doc/libs/1_57_0/libs/graph/doc/index.html
>
> It also has a substantial parallel graph library (actually, it's more a
> "distributed" graph library):
>
> http://www.boost.org/doc/libs/1_57_0/libs/graph_parallel/doc/html/index.html
>
> All the things you mentioned are already present in those libraries.
>
> I suggest that you look into those libraries to see what is already there,
> and what you might be able to add to it. Otherwise, there isn't much point
> in having you do a GSoC that reinvents the wheel and doesn't provide
> anything new to Boost.
>
> Cheers,
> Mikael.
>
>
> --
> Sven Mikael Persson, M.Sc.(Tech.)
> PhD Candidate and Vanier CGS Scholar,
>
> Department of Mechanical Engineering,
> McGill University,
>
> _______________________________________________
> Unsubscribe & other changes:
> http://lists.boost.org/mailman/listinfo.cgi/boost
>


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk