RE: Notes meeting Wed Feb 14 (fwd)

[Elton Smith <elton@jlab.org>]

---------- Forwarded message ----------
Date: Wed, 14 Feb 2001 17:18:46 -0500
From: Larry Dennis <dennisl@csit.fsu.edu>
To: Elton Smith <elton@jlab.org>
Cc: Elliott Wolin <wolin@jlab.org>, Larry Dennis <dennisl@csit.fsu.edu>
Subject: RE: Notes meeting Wed Feb 14 (fwd)

Hi Elton,

   I'm not sure if I am on that mailing list or not.  I get all the e-mails
from Alex and others -- but those seem to come from IU.

   I think your notes capture the essence of the discussion just fine.  I am
convinced that we need a way to get groups to accept responsibility for
software subsystems.  If we follow the lead of the HEP and RHIC computing
groups and put some of the Hall D computing resources at remote sites, then
I think we will be better able to attract strong groups to the software
projects, but also help get the organizational infrastructure in place as to
how we will conduct our computing and will be better able to take advantage
of non-standard funding opportunities (International funding, University
resources, DOE and NSF ITR initiatives).

	The way I think this might work for simulations is that we would ask two or
three groups to take responsibilty for the simulations.  One group might
focus on the straw tubes and beamline, another on barrel and start counter,
another on cerenkov and detector supports (I'm not  sure what exactly is the
right way to divide this up into two or three groups.)  It would be the
responsibilty of these groups to develop the simulation software, and to
design, deploy and manage the simulation farms.  To do this they would be
expected to pursue at least partial University support for this effort.  In
return they would get the construction funding for their part of the
computing infrastructure and perhaps we could let DOE know that they are
prime candidates for post docs.  I think this is precisely the model being
developed for the LHC detectors (on a much smaller scale).

   Unfortunately, I left my notes on budgets at home this morning, my
recollection was that we needed something like $700 - $800 K to develop
customized computing facilities at several remote sites.  I will get them
into the computer and send them to you and Elliott tonight.

Larry Dennis