Fw: Regarding Participation in GSOC-2013 for "Fast Linear Algebra over Extension Fields"

24 views
Skip to first unread message

Burcin Erocal

unread,
Apr 30, 2013, 4:03:05 AM4/30/13
to lela-...@googlegroups.com, tejesh...@gmail.com
Hi Tejesh,

thank you for your interest.

communication about this project is taking place on the
lela-...@googlegroups.com mailing list. I am forwarding your message
there. Answer will come later.

Cheers,
Burcin


Begin forwarded message:

Date: Tue, 30 Apr 2013 09:42:12 +0530
From: Tejesh Doddikoppad <tejesh...@gmail.com>
To: bur...@mathematik.uni-kl.de
Cc: Martin Rääk <martinr...@googlemail.com>
Subject: Regarding Participation in GSOC-2013 for "Fast Linear Algebra
over Extension Fields"


Hello Sir,

I am Tejesh Doddikoppad, masters student from Manipal University, India.
Although, I am applying bit late due to my exams but I managed to study
this project over the last week and I think I am able to contribute to
lmonade community.

As much as I understand ,the goal of the project is to:


- Write low level linear algebra routines like Matrix-Multiplication,
addition using extension in finite field approach.These low level
routines are essential part of many computational algebra
algorithm and making these routines fast increases the
performance of the whole algorithm.
- Writing the routines in the modular fashion.

The approach of making these routines (like vector addition) fast is to
break the routines into multiple small routines (pointwise addition) and
finally merge them. I have worked on matrix multiplication using similar
kind of approach. In which I broke the whole matrix into smaller blocks
and then executed matrix multiplication on these blocks on different
parallel computing units. I also implemented the same routine with
other approaches like row wise matrix multiplication and point-wise
matrix multiplication. I am attaching the codes with this mail.

To give you background about myself. Currently I am working at ST
Microelectronics as a Intern Student and working on 3D video compression
and free view point (FVV) synthesis. I worked here with C/C++. I have
also exposure of parallel computing domain. My Internship period will
be over by next week and I have no commitments for this summer.

I am very interested in doing this project and have enough time to
complete the project during the summer.

Please reply to this mail for further communication.

Thanking you,

Tejesh

Original html part
MATRIX MULTIPLICATION.zip

Martin Albrecht

unread,
Apr 30, 2013, 5:02:02 AM4/30/13
to lela-...@googlegroups.com, tejesh...@gmail.com
Hey Tejesh,

did you search the archives of lela-users, we discussed the project here a
bit. If you have more questions feel free to ask here.
Cheers,
Martin

--
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF
_www: http://martinralbrecht.wordpress.com/
_jab: martinr...@jabber.ccc.de
Reply all
Reply to author
Forward
0 new messages