[Biojava-dev] Fyi: Client side high performance computing project using Biojava

[Raphael André Bauer]

Hi,

OUR IDEA:
We want to make web browsers and their javascript engines to act as
part of the largest computing grid of the world. If you think of
projects like folding at home or seti at home you know what we are talking
about. But our approach is much simpler and way more powerful: Go to a
website and you are participating in a huge grid - computing
interesting (scientific) stuff. No software installation required.

CURRENT STATUS:
We are bioinformaticians from Berlin and San Diego. Naturally we are
using scientific tasks from the field. Our first success was to
crosscompile a package called biojava (biojava.org) into javascript
using Google Web Toolkit. We are now able to calculate protein
sequence alignments on the client side. The tasks are distributed by
the server. Check out our live demo at:
http://js-cloud-computing.appspot.com/. First results and detailed
information can be found under:
http://code.google.com/p/js-cloud-computing/

WHERE TO GO FROM HERE:
First of all this mail is a simple fyi. We would be very happy to get
in touch with people which are interested (check out our mailing list
at http://groups.google.com/group/js-cloud-computing-dev). Feedback is
great - and of course our source code is freely available at google
code ( http://code.google.com/p/js-cloud-computing/source/browse/).

The next steps in this project will include the cross-compilation of
the structure-package of the biojava package. using GWT. The idea is
to align proteins in 3D space on the javascript client side :)

We are looking forward hearing from you - and sorry for crosspostings :)


Cheers,

JSCC Core Team
Nico, Güttler, Marcus Schroeder, Raphael Bauer, Andreas Prlić