On Fri, 14 Jan 2005, Burke, S (Stephen) wrote:
> Nothing major then :) I seem to remember when I first joined EDG I started
> by trying to understand what an SE was supposed to be; I'm not sure I've
> ever managed it ... part of the problem is that sites just think "we're
> supposed to have an SE", so they put one in even if it's just a normal
> machine with a 100 Gb disk. As you say, given good networking there's no
> strong reason for that, sites can easily point their close SE somewhere else
> - as NIKHEF have apparently done, although Sara isn't entirely a different
> site.
The problem is that requires WNs to have IP connectivity to the world at
large, a requirement that we at Brunel would like to avoid.
My hope was that the middleware would evolve in such a way that the job on
the WN can upload and register the results to the local cache-SE, request
the local cache-SE to *move* the data to whichever mass-storage-SE(s)
the experiment thinks is appropriate and then terminate immediately,
returning some handle to the move-request to the experiment's submission
framework so that progress can be tracked or the move-request modified if
there's a problem somewhere. (Note how the "RLS" service is accessed
through the SE, too)
Of course, this requires a much heavier SE than at present...
(I suppose strictly the cache-SE doesn't have to be local, but you get
the idea...)
Henry
--
Dr. Henry Nebrensky [log in to unmask]
http://www.brunel.ac.uk/~eesrjjn
"The opossum is a very sophisticated animal.
It doesn't even get up until 5 or 6 p.m."
