California Institute of Technology
Dr. Daniel I. Meiron
Applied Mathematics 217-50
Firestone Laboratory
Pasadena, CA 91125
Tel: (626) 395-4563
Fax: (626) 578-0124
e-mail: dim@its.caltech.edu
November 1, 2003
Prof. Harvey Newman
247 Lauritsen
M/C 256-48
Caltech
Pasadena, CA 91125
Dear Harvey
I am writing to add the support of CACR for the proposed work described in your revised proposal entitled “A
Global Grid-Enabled Collaboratory for Scientific Research”
I am currently serving as the Acting Director of Caltech’s Center for Advanced Computing Research. As you know
CACR is dedicated to the pursuit of excellence in the field of high-performance computing, communication, and
data engineering. Major activities include carrying out large-scale scientific and engineering applications on parallel
supercomputers and coordinating collaborative research projects on high-speed network technologies, distributed
computing and database methodologies, and related topics. The proposed work is in perfect alignment with the
CACR mission.
We have enjoyed a strong and successful partnership between HEP and CACR since 1996, on a variety of projects
related to Petascale database access and analysis, Grid systems, and most recently high performance networks. The
development of a globally scalable Grid-enabled Collaboratory for scientific work is a natural and important next
step that could be of substantial mutual benefit as regards CACR’s participation in the Teragrid project. CACR
stands ready to aid this work in a variety of ways such as support for high performance computational facilities and
networking. If I can be of any assistance in this regard please do not hesitate to contact me.
Sincerely,
Dan Meiron
Acting Director, CACR, Caltech
110, rue O?Connor Street, 4
e
Øtage / 4
th
Floor, Ottawa, Ontario, Canada K1P 5M9
Tel./TØl.: (613) 943-5454
•
Fax./TØlØcopieur: (613) 943-5443
•
e-mail: info@canarie.ca
•
www.canarie.ca
C A N A D A ? S A D V A N C E D I N T E R N E T D E V E L O P M E N T O R G A N I Z A T I O N
L E F E R D E L A N C E D E L ? I N T E R N E T ? V O L U ? A U C A N A D A
Dr. Shawn McKee
University of Michigan Physics Department
500 East University
Ann Arbor, MI, 48109-1120
November 1, 2003
Dear Shawn:
I am delighted to be able to express my support for the "Global Grid-Enabled Collaboratory for
Scientific Research (GEC SR)". Since the early days of such concepts as collaboratories and
grids, the high energy physics community has always been seen as leader in advancing these
technologies and concepts.
The ability to harness the expertise of such a collection of institutions combined with existing
tools, and developmental abilities in collaborative systems and working methods, Grid and
distributed software development and optimization, high performance networking, and related
areas, and to work with leading computer scientists as well as major HENP laboratories and the
physics community, I am confident will ensure the program’s success and effectiveness.
I believe GEC SR could have a transformative impact on our research in the fields of High
Energy and Nuclear Physics, as well as many other fields within and beyond the bounds of
research. The work program outlined in the proposal is indeed essential for the success of our
current and next-generation global scientific collaborations. If these developments are applied
broadly, they will have significant benefits to research in many fields, to education in the form of
new learning environments, and more broadly to enable collaboration and collaborative work
across networks on a global scale.
As you know CANARIE and the Canadian research community has extensive interest in the
development of these tools as well and we would look to future collaboration with your team
upon successful award funding.
Sincerely
Bill St. Arnaud, P.Eng.
Senior Director Network Projects
CANARIE Inc
University of Illinois
at Urbana-Champaign
National Center for Supercomputing Applications
605 East Springfield Avenue
Champaign, IL 61820
USA
Daniel A. Reed
(217) 244-0078
(217) 244-8195
FAX
reed@ncsa.uiuc.edu
November 4, 2003
Dr. Shawn McKee
Department of Physics
University of Michigan
Ann Arbor, Michigan 48109
Dear Shawn:
I am writing in support of your NSF proposal for an integrated collaboration environment, known
as "A Global Grid-Enabled Collaboratory for Scientific Research (GEC SR)." Emerging
approaches to computational science require real-time computing support, large-scale analysis
and management of terabytes of observational data. Concomitantly, they require mechanisms
for physically and temporally distributed groups, often containing thousands of collaborators, to
jointly coordinate data analysis and experimentation. I believe the proposed work will help
simplify distributed collaboration and interaction, a critical component of national and international
scientific teams.
Traditional collaboration and teleconference systems impose a simplistic and constraining model
of distributed interaction that is not well suited to the dynamic, and often informal information
sharing and discussion central to scientific research. Your proposal addresses precisely these
issues, promising to create a flexible collaboration infrastructure that is tested and validated on an
international scale.
I am also delighted to see that the proposal includes a well-balanced mix of application and
computing researchers who all have an outstanding track record of collaboration and software
development. The grid infrastructure research and developments of the GriPhyN and iVDGL
projects have had profound impact on the international physics community and its ATLAS and
CMS projects.
This work is also reflective of NCSA and the Alliance’s broad vision of a distributed Grid
infrastructure based on commodity hardware and widely distributed community software,
together with scalable distributed collaboration systems. Based on this vision, as part of the NSF
TeraGrid, NCSA will deploy 10 teraflops of Linux cluster computing capability and over 500 TB of
storage, connected by a 40 Gb/s transcontinental network to additional TeraGrid resources and
enabling distributed data management, computation, and visualization. We look forward to
working you on high-energy physics data analysis atop the TeraGrid.
In addition, the Alliance Scientific Workspaces of the Future (SWOF) expedition is developing a
collaboration infrastructure that combines the next generation Access Grid with distributed
information sharing, visualization and collaboration support. This expedition is in the same spirit
as your proposed work, and I also look forward to fruitful collaborations between the GEC SR
team and the Alliance.
November 4, 2003
In summary, I believe the vision outlined in this proposal exemplifies the future of scientific
computing: distributed, data driven, and community based. It has my enthusiastic
recommendation and support.
Sincerely,
Daniel A. Reed
Edward William and Jane Marr Gutgsell Professor
Director, National Center for Supercomputing Applications
Director, National Computational Science Alliance
Chief Architect, NSF ETF TeraGrid
Columbia University
DEPARTMENT OF PHYSICS NEVIS LABORATORIES
136 South Broadway
P.O. Box 137
Irvington, NY 10533
Tel. 914-59 2809
Fax 914-591 8120
6 November 2003
Re: The Global Grid-Enabled Collaboratory for Scientific Research
Proposal
To Whom It May Concern:
Recent discussions of the methods for actually operating a large physics experiment with
many people from around the world on shift and fully participating in the monitoring,
evaluation and decision making have reinforced the urgent need for development of tools
for grid-based collaboration in real time. The time-zone problem that ordinarily is
consider a problem in global collaboration is transformed into an advantage when shifts
around the clock must be mounted, each incorporating a number of experts in different
aspects of the collaboration. It is quite unrealistic to imagine bringing to one spot all the
staff needed to operate such a facility seven days a week and nearly every day in the year.
In the case of the Large Hadron Collider experiments, where operation of the detectors
begins in less than three years, the need to get the tools in place is very urgent, but no
doubt there are many other operations that need just the same sort of tools.
Several years ago, when the development of these methods started at a small scale, I
found the effort that has now presented this proposal addressed exactly the need that has
seemed to me the biggest impediment to global collaboration, the need for tools that
allowed much more comprehensive interaction among different kinds of associations of
collaborators than our present video and phone conferences. It is a commonplace
observation that the need for travel to a central site has not decreased as much as we had
hoped. The hallway and cafeteria interactions at the key nodes of a collaboration are
found to give such high payoff that it makes the arduous and expensive travel
worthwhile. Even in a conference around a table, the video version is very inferior to a
real one in many cases. The need to exchange large amounts of information in a parallel
way between individuals and small and large sets of collaborators is also important.
The solution of this problem evidently requires a lot of work. The GECSR Proposal
brings together a number of groups who represent recognized experts on this kind of
problem and other groups skilled in computer science and physics analysis. They
acknowledge the difficulty of this task, and are bringing to bear a number of different
skills and tools in a phased, iterative approach. We now believe that they are in a
position to achieve success. I trust that they will be supported to bring a solution to this
important problem in the short time before we have to start our operations.
Yours sincerely,
William J. Willis
William J. Willis
Eugene Higgins Professor of Physics
U.S. ATLAS Project Manager
Internet2
3025 Boardwalk, Suite 200
Ann Arbor, MI 48108
(734) 913-4250
(734) 913-4255 (fax)
www.internet2.edu
November 6, 2003
Harvey Newman, Ph.D.
Professor of Physics
California Institute of Technology
Mail Code 256-48
1200 E. California Blvd.
Pasadena, CA 91125
Dear Harvey:
On behalf of the University Corporation for Advanced Internet Development (UCAID) and its
Internet2 project, I am pleased to write in strong support of the ITR proposal entitled A Global
Grid-Enabled Collaboratory for Scientific Research that you are submitting to the National Science
Foundation.
This proposal represents an extremely effective collaboration led by California Institute of
Technology and the University of Michigan, in concert with the University of Texas, University of
Maryland, University of Iowa, Florida International University, Fermilab, the Lawrence Berkeley
Laboratory and other institutions. Together, this dynamic group will develop a next-generation
collaborative system for research, harnessing multi-institutional expertise, technological tools, and
developmental capabilities.
The proposals strengths include its:
Effective collaborative infrastructure linking leading computer scientists from cutting-edge
research universities, researchers from major HENP laboratories, and the physics research
community as a whole;
Unique combination of joint efforts addressing collaborative systems and working methods,
Grid and distributed software development and optimization, and high-performance
networking;
Unprecedented focus on collaborative Grid-enabled and other analysis tools to facilitate data-
intensive analysis involving worldwide-distributed resources within a global research
community; and
Enhancement of the cooperative and productive HENP collaborative infrastructure that will
likely serve as a model for other multi-organizational efforts.
This proposal represents a partnership between Grid technologies and the HENP model of
collaboration. It is likely to create models for many other scientific areas involving current and
next-generation data-intensive global collaborations. The proposal’s research outcomes are also
Page –2-
likely to have a significant impact on education through the creation of new learning environments,
and on other multi-institution organizational infrastructures to enhance investigative interaction
across global networks.
We support this proposal with the strongest degree of enthusiasm.
Sincerely,
Douglas E. Van Houweling
President and CEO, Internet2
Geoffrey Fox
Community Grids Laboratory
Indiana University
Bloomington IN
Dear Harvey:
It is my pleasure to endorse the GECSR project submitted to the NSF Physics
at the Information Frontier program. Such collaborative capability is
essential to realize the promise of Grids which upto now have concentrated
on largely asynchronous saring of compute and data resources. It is a
central part of the idea that Grids support all aspects of a virtual
organization including people as well as electronic resources. The lead
institutions are highly qualified: CHEF from Michigan and VRVS from Caltech
are leading projects in their areas of collaboration. Caltech through CMS
has played a major role in projects like GriPhyN, iVDGL, PPDG and recent
work on the development of a Grid-enabled Analysis Environment using Web
services. These have pioneered critical demonstrations that Grids are
important.
I am already working with CHEF as part of the MSF Middleware Initiative and
I would be happy to further work with your team and contribute ideas and
technologies from our GlobalMMCS (Multi Media Collaboration system) project
with Beihang University which focuses on Collaborative Web Services and
Collaboration as a Web Service.
Geoffrey
Dear Harvey,
With a great interest we are learning about the preparation of a
proposal on "Global Grid-Enabled Collaboratory for Scientific Research "
by Caltech and the University of Michigan. This proposal is going the
same direction as an European initiative on "Enabling Grid for e-science
in Europe" to be submitted to the 6 FP of EU this spring.
Science brings us new challenges all the time, which we have to
respond; bio-chemistry, earth observation, astrophysics, particle
physics are just a few fields, which already now produce enormous amount
of data, requesting storage, fast processing and an easy access. In a
near future we will face even greater demands, as LHC particle physics
experiments will produce PetaBytes of information per year.
These experiments are the result of the world-wide
collaborations, which ask for world-wide data analysis. The task is not
easy, however a steady progress on technologies, and new ideas, give a
chance to find a solution. The Grid concept, which was born in the
United States, belongs to those.
Last years several Grid projects have been launched on both
sides of the Atlantic to address the issue of data processing for the
future LHC experiments: GryPhiN and iVDL in the States, DataGrid and
CrossGrid in Europe. The LHC Computing Grid project, coordinated by CERN
has started. We think that these projects need attention and support,
and coordination on a global scale. In this respect, EGEE initiative,
and your proposal, go very much along, and we expect a fruitful
cooperation.
As the participants of the CrossGrid project, concerned about
particle physics applications, and as members of ATLAS and CMS
communities, we express our interest and support to your initiative. We
hope that all these efforts, will bring results which will allow us to
succeed in a global scale computing needed for LHC.
Jesus Marco
(CrossGrid testbed leader, CMS)
Michal Turala
(CrossGrid project coordinator, ATLAS)
Professor John Huth
236 Lyman Laboratory
Harvard University
Cambridge, MA 02138
617 495 8144
617 495 0416 (fax)
huth@physics.harvard.edu
7 Nov, 2003
Shawn McKee
Physics Department
500 East University
Ann Arbor, MI 48109-1120
As the Associate Project Manager for Physics and Computing for the U.S. ATLAS
Project, I want to strongly endorse the proposal entitled ?The Global Grid-Enabled
Collaboratory for Scientific Research?. The effectiveness of global collaborations, such
as those engaged in the Large Hadron Collider (LHC) depends on tools to enable remote
users to participate in a meaningful way. The research and proposed here directly
address this issue.
Historically, collaborators located at the central location of an experiment have had an
undeniable advantage over remote collaborators. With new grid technologies becoming
available, and high bandwidth links, it is clear that there is a substantial opportunity to
empower remote users to become highly effective collaborators. The group participating
in this proposal already has substantial experience in collaboratory efforts. I believe that
they are in an excellent position to understand the issues in remote collaboration and have
a sensible approach to the problem. My collaboration stands to benefit highly, if they
realize the goals set forth. I strongly support this proposal.
Sincerely,
John Huth
f
F
e
r
m
i
l
a
b
L
o
t
h
a
r
A
.
T
.
B
a
u
e
r
d
i
c
k
P
r
o
j
e
c
t
M
a
n
a
g
e
r
U
.
S
.
C
M
S
S
o
f
t
w
a
r
e
&
C
o
m
p
u
t
i
n
g
F
e
r
m
i
N
a
t
i
o
n
a
l
A
c
c
e
l
e
r
a
t
o
r
L
a
b
o
r
a
t
o
r
y
P
.
O
.
B
o
x
5
0
0
•
M
S
2
3
4
B
a
t
a
v
i
a
,
I
L
•
6
0
5
1
0
•
P
h
o
n
e
6
3
0
-
8
4
0
-
6
8
0
4
•
F
a
x
-
2
7
8
3
November 10, 2003
Re:
Proposal
“PIF:
Global
Grid-Enabled
Collaboratory
for
Scientific
Research”
To
Whom
It
May
Concern:
As
leader
of
the
U.S.
CMS
Software
and
Computing
project
I
wish
to
express
my
support
for
the
PIF
proposal
to
the
NSF
by
Prof.
Harvey
Newman
of
Caltech,
Dr.
Shawn
McKee
of
U.Michigan
and
a
group
of
computer
scientists
and
physicists
at
several
Universities,
collaborating on the CMS and Atlas experiments at the Large Hadron Collider. They propose to
develop new technologies and build an operational Grid-Enabled Collaboratory for scientific
research.
The LHC at CERN in Switzerland will provide one of the most exciting, challenging, and relevant
pieces of science over the next 20 years. U.S. scientists are eagerly looking forward to harvest the
scientific yield of many years of preparations and a half billion dollar U.S. investment, when in a
few years the LHC starts to collide particles. The LHC’s scientific and technical challenges are
daunting, but as importantly there will be large cultural and organizational issues to solve when
several
thousand
scientists
around
the
world
will
start
to
collaborate
on
a
single,
multi-purpose
facility with a very broad scientific program.
The GEC SR project sets out to provide an environment for collaboration between science
communities. This could very well be the cornerstone for the success of the LHC scientific
program,
specifically
for
the
U.S.
university
groups
that
are
separated
from
the
experiment
