Re: Distributed networks?

["Robert J. Bradbury" <bradbury@aeiveos.com>]

On Thu, 26 Aug 2004, Chris Albertson wrote:

> TASS is on the "wrong" end of this.  You would have to
> send out megabytes of data (several full sized FITS images)
> and the processing takes only about 60 seconds.
> The ratio of data transmitting to data processing is
> not good.

I could propose taking the processors to the data --
(one can now get 4-processor boards for AMDs that
are relatively cheap).  I was looking however at
what approaches are out there.

Since one presumably has to match the images to
each other in *any* case I don't see the problem
in subdividing an image by 4, 8, 16, etc. for
processing.  You allow enough overlap and stay
away from the edges where required so that you
avoid the problem of points getting split.
Astronomy ought to be way past the point where
the problems with merging/splitting images are
that significant.  Heck a few years back I even
had a small scanner/camera from HP that would
scan book pages and knit them together without
any difficulty.  This isn't rocket science... :-;

(Or maybe I'm missing something about what is going
on here...)
Robert