Achieving

1
Achieving Sustainable Stewardship
State of ASCAddress to the 2007 Principal
Investigator MeetingDimitri KusnezovDirector,
Office of Advanced Simulation Computing
2
A busy year
Scientific experts express doubt about Bush
administration's new warhead plan
ASSOCIATED PRESS 559 p.m. February 19, 2007 S
AN FRANCISCO A panel of scientific experts
expressed skepticism about a Bush administration
plan to revamp the nation's nuclear weapons
arsenal with a warhead now on the drawing board.
The panel, composed mostly of former Energy Depa
rtment and nuclear-weapons lab executives,
refused to endorse the proposal, saying that
there is inadequate evidence that the current
stockpile is deteriorating as the administration
claims. The uncertainty about the warheads' lifes
pan is serious and it's legitimate, but it's not
yet empirical, said the panel's chairman,
physicist C. Bruce Tarter, a former director of
Lawrence Livermore National Laboratory. The lab
has bid to oversee the development of the
proposed new warhead. Los Alamos National Laborat
ory in New Mexico is the other facility competing
for the lead role in the development of the new
bomb, dubbed the Reliable Replacement Warhead, or
RRW. The Bush administration had planned to annou
nce a decision several weeks ago on which lab's
design would prevail, but unknown factors have
delayed the decision. The experts presented their
doubts before the annual meeting of the American
Association for the Advancement of Science. They
found many of the claimed benefits of RRW vague
and uncertain, and said they will not be felt
until some time in the distant future.
If there aren't numbers for costs or schedules,
how do you know it's better than what you're
doing? Tarter wondered. Both labs have described
their proposals for the new warhead as
conservative blueprints meant to assure
reliability without violating a moratorium on
full-scale nuclear testing in place since 1992.
That inability to test was one of the factors
that concerned the scientists. Certifying a new
nuclear explosive package remains an unproven
technical feat, said Raymond Jeanloz, a panel
member and University of California, Berkeley
planetary scientist. But Gen. James Cartwright, c
ommander of the U.S. Strategic Command, which
oversees U.S. nuclear forces, said RRW is vitally
important. If the policy is to have nuclear weap
ons, the policy ought to be to make them as
secure as possible, as safe as possible. Anything
less is irresponsible, he said.
The main objectives of FP7 Specific programmes
Knowledge lies at the heart of the European Unio
n's Lisbon Strategy to become the "most dynamic
competitive knowledge-based economy in the
world". The 'knowledge triangle' - research,
education and innovation - is a core factor in
European efforts to meet the ambitious Lisbon
goals. Numerous programmes, initiatives and
support measures are carried out at EU level in
support of knowledge. The Seventh Framework Pr
ogramme (FP7) bundles all research-related EU
initiatives together under a common roof playing
a crucial role in reaching the goals of growth,
competitiveness and employment along with a new
Competitiveness and Innovation Framework
Programme (CIP), Education and Training
programmes, and Structural and Cohesion Funds for
regional convergence and competitiveness. It is
also a key pillar for the European Research Area
(ERA).
Saturday, February 17, 2007 Hybrid Warhead
Panned By John Arnold Journal Staff Writer   
A senior Los Alamos National Laboratory weapons
scientist says a program to develop a new kind of
nuclear warhead will fail if the government takes
a "frankenbomb" approach to its design.   
Weapons designers at LANL and California's
Lawrence Livermore National Laboratory are
competing to design the Reliable Replacement
Warhead, which would replace existing warheads in
the nation's aging Cold War-era arsenal.    John
Pedicini, LANL's design team leader, believes
LANL won the competition based on the technical
merits of its design. But Pedicini wrote on a Web
log that following the technical competition,
federal nuclear weapons officials asked the rival
labs to come up with a "plan of cooperation,"
including a hybrid warhead design that would
include elements from both teams' plans.    "A
hybrid design by inexperienced personnel, managed
by committee, is not the best approach... The
best appellation I have seen for such an approach
is 'frankenbomb,' '' Pedicini writes in an entry
posted on the site "LANL The Corporate
Story."    Noting that he was writing as a
private citizen and not as a representative of
the lab, Pedicini said that he has been told by
"multiple highly placed sources, with ranks into
sub-cabinet level" that LANL won the technical
competition.    He added that there are some
features of the Livermore design that are an
advance over his own team's plan, and that he
would incorporate those advances, should LANL win
the competition.    "If this is what is meant by
hybrid, then the outcome would be good," said
Pedicini, who did not return phone or e-mail
messages seeking additional comment.    Weapons
designers were asked to come up with plans for a
replacement warhead that could be deployed
without underground nuclear testing. The teams
submitted their designs last year to the Nuclear
Weapons Council, an inter-agency panel of
Department of Energy and Pentagon officials who
decided last fall that the Reliable Replacement
Warhead program should move forward.    The Bush
administration's budget proposal for fiscal year
2008 calls for tripling program funding to 88
million.    But the weapons council has yet to
name a lead design team, fueling speculation
among lab workers and observers about the
competition's outcome.    "There are a lot of
rumors out there, and there are a lot of people
who think they know what they're talking about.
As I understand it, it's a very small group of
people that really do know the situation and are
discussing it, and that's the Nuclear Weapons
Council," said National Nuclear Security
Administration spokesman Bryan Wilkes.    LANL
spokesman Kevin Roark said that Pedicini's
comments were not made on behalf of the lab, but
he declined to comment further.    Pedicini is
the second senior LANL weapons designer to speak
publicly this month, without lab management's
permission, about the Reliable Replacement
Warhead and nuclear weapons policy.    Joe
Martz, LANL's Reliable Replacement Warhead
project leader, told the San Francisco Chronicle
that the U.S. should consider a nuclear weapons
policy that would eventually eliminate the
country's nuclear arsenal.    He also believes
that it's unlikely LANL will win the Reliable
Replacement Warhead competition, due to security
lapses highlighted during a congressional hearing
last month, according to a Chronicle article
published Tuesday.    In his Web log post,
Pedicini calls the Jan. 30 congressional hearing
"a wild card whose influence I cannot gauge."   
The Reliable Replacement Warhead is a key
component of NNSA's plan to consolidate and
modernize the nation's nuclear weapons
complex.    Supporters say the new warhead
design will be easier and safer to maintain and
will offer more protection against unauthorized
use. It will also allow the government to
dismantle old weapons and ultimately reduce the
size of the stockpile, NNSA officials have
said.    However, critics say that programs to
maintain and extend the life of the existing
nuclear weapons have been successful. Going
forward with a new warhead design is unnecessary
and will undermine global nonproliferation
efforts, they contend.
Europaeischer Forschungsrat ab 2007Mit dem
Start des 7. EU-Forschungsrahmenprogramms (7.
FRP) 2006 wird auch der neu einzurichtende
European Research Council seine Arbeit aufnehmen.
Die Gruendung eines solchen Europaeischen
Forschungsrats geht auf einen Vorschlag der
Europaeischen Kommission innerhalb des Entwurfs
fuer das 7. FRP aus dem Jahre 2005 zurueck. Ziel
des European Research Council wird sein, junge
Nachwuchswissenschaftler zu foerdern und
bahnbrechende Grundlagenforschung in allen
wissenschaftlichen und technologischen Feldern,
auch in den Sozial- und Geisteswissenschaften, in
Europa zu unterstuetzen. Dazu steht ihm ein
Budget von 7 Mrd. (2007-2013) zur Verfuegung.
Ein unabhaengiger Wissenschaftlicher Rat aus 22
Fachmitgliedern prueft die Antraege und
entscheidet ueber die Foerderungen bzw.
Stipendienvergabe.Foerderlinien,
Begutachtungsverfahren und Dokumente zum ERC
unterhttp//ec.europa.eu/erc/index_en.cfm
3
Another Remarkable YearLeadership in Scientific
Computation

• Integrated activities across campaigns led to
  successful completion of FY06 Level 1 milestone
  on Pit Lifetime Assessment
  
• Delivered on 10-year, 100 teraflops commitment
  with ASC Purple and now Red Storm
  
• New understanding of primary physics achieved in
  first month of operation on Purple (genesis of
  Thermonuclear Burn Initiative)
  
• Record sustained computing performance on BG/L
  (200 teraflops) applied to Pu aging
  
• ASC technology enabled feasibility study of first
  RRW design, solution of physics and engineering
  designs electronically coordinated between labs
  and manufacturing sites
• RRW design process provides glimpse into the
  complex Responsiveness
  
• Application of casting code to designing RRW pit
  manufacturing processes
  
• Increased confidence in our national nuclear
  security through improved predictivity of the
  physical processes in stockpile systems (W76-1,
  W88 MAR, B-61, Annual Certification and
  Assessment,)
• Significant improvements in accuracy, usability,
  and time to solution of modern 3D codes
  
• Two new product lines, Red Storm and BG/L
  consistent with the Presidents American
  Competitiveness Initiative (Supporting Office of
  Science siting of leadership class machines at
  Oak Ridge and Argonne National Laboratories)
• Rapid response to emerging threats (DPRK)
• Remarkable progress in large scale simulations at
  Alliance Centers

4
ASC Complex 2030
Concepts basic to complex 2030 transforming
the stockpile, cost-effectiveness of the
enterprise, integration and consolidation, and
driving the ST base. Complexity will increase
during transformation as the stockpile diversity
increases and hands-on test experience decreases.
For the first time in the history of nuclear we
apons, our nuclear deterrent may rely upon
devices that, though carefully compared with the
underground test data, have not been tested in
their nuclear regimes.
Increased stockpile diversity
Increased confidence in warhead designs and
demonstration of a responsive infrastructure will
enable a reduction in total stockpile size
5
LANL Designers Test Experience
6
When validation replaces calibration

• Desired state Confidence in science-based
  simulations exceeds our confidence in simulations
  calibrated by underground test data

7
Thermonuclear Burn Initiative
Back to basics Leadership science Complex mult
i-scale Frontier computational science Nationa
l security 2012 commitment No milestones in
tellectual headroom
8
ASC Roadmap
www.nnsa.doe.gov/ASC
9
Predictive Capability Framework (PCF) integration
across NNSA central to Complex 2030
Integration of ASC, DSW, Science, Engineering,
and ICF Campaigns modeled on ASC Roadmap
High-level goals from ASC Roadmap
Required increment in capability
Incorporates existing roadmaps and strategies
(e.g., Primary Assessment Strategy)
Defines by program needs the required capabiliti
es (e.g., ongoing LEPs define current needs)
Provides close integration between experimental
and computational activities
Tier 1 and 2 accomplishments
Goal Deliver improved and increasingly
predictive capabilities to support stockpile
needs and enable transformation of the complex
into a responsive infrastructure
10
2007-08 ASC Roadmap Outcomes and Outputs

• Focus area 1 Address national security
  simulation needs
  
• 2008 Deliver a national code strategy (Ken
  Alvin)
  
• Focus area 2 Establish a validated predictive
  capability for key physical phenomena
  
• 2007 Thermonuclear Burn Initiative
  collaboration
  
• 2012 horizon
• 2008 Deliver realistic plutonium aging
  simulations
  
• BG/L enabled
• Focus area 3 Quantify and aggregate
  uncertainties in simulation tools
  
• 2008 Deliver a national verification and
  validation strategy (Karen Pao)
  
• 2008 Assessment of major simulation
  uncertainties
  
• Focus area 4 Provide mission-responsive
  computational environments
  
• 2007 Initiate a new National User Facility model
  for capability supercomputing
  
• 2008 Implement seamless user environments for
  capacity computing

11
Defense Programs Funding HistoryFY 2001 FY 2008
in Billions (Then year dollars shown in FY 2007
Dollar)
12
FY 2008 Priorities
A slide from last weeks briefings to House and
Senate staff

• Application of ASC codes to support
  certification, nonproliferation,
  attribution/forensics activities, and emergency
  response
• Identify and characterize special nuclear
  material threat materials and devices
  
• Enable forensic analysis of post-explosion radio
  nuclei debris
  
• By FY 2009, develop, implement, and validate a
  suite of physics-based models and high-fidelity
  databases in support of the full operational
  capability of the Defense Threat Reduction Agency
  National Technical Nuclear Forensics program.
• Computational surrogate for nuclear testing
  weapons behavior moving towards predictivity
  
• Emphasis on quantification of confidence in the
  results of weapon simulation codes in order to
  inform Quantification of Margins and
  Uncertainties (QMU) certification and assessment
  methodologies
• Establish a national UQ forum from various
  verification and validation communities to
  establish common framework for addressing UQ
  challenges and conduct peer reviews
• By FY 2009, develop a national verification and
  validation program strategy, which incorporates
  the tenets of uncertainty quantification.
  
• Second year of Thermonuclear Burn Initiative
  (TBI) the 2012 moon shot to understand a
  leading source of uncertainty.
  
• High-performance computing as a key enabler to
  the broad NNSA mission
  
• Sequoia - platform targeted at development of
  uncertainty quantification (UQ) and validation
  critical to development of certification
  methods.
• Coordinated strategy with other agencies

13
ASC Leading Change

• We have had broad impact
• Thermonuclear Burn Initiative ? Boost initiative
• Roadmap ? PFC
• National User Facility model
• nWBS ? Defense Programs nWBS
• Looking Ahead
• Computing as a core competency beyond weapons
  certification
  
• NN, DTRA, Office of Science,
• RRW futures
• Defining the national agenda for VV methods in
  complex computations
  
• Software Strategy
• Complex 2030 Challenges Computing consolidation
• Complex transformation within a level budget
  profile
  
• Two National Academy of Sciences Studies One
  JASON
  
• Science Technology and Engineering 2030 Roadmap

14
The next 3 5 years
Building on our successes Defining the frontier
where we lead

• Demonstrate progress toward reduced phenomenology
  in the simulations with credible uncertainties,
  based on robust code verification and validation,
  in the prediction of device behavior.
• Apply programming models to new architectures
  that make best use of their potential computing
  power for full weapons codes.
  
• Through a National Code Strategy, ensure that the
  next generation of major physics and engineering
  codes reflect the national security needs of the
  United States, including attribution and secure
  transportation.
• Incorporate Uncertainty Quantification as a
  formal step in the analysis process and
  application of ASC simulations.
  
• Align the ASC workload with the ASC Roadmap focus
  areas to meet the associated targets. In an
  environment of steady to decreasing budgets, this
  will mean prioritizing current workload with new
  efforts.
• Continue operation of a capability platform as a
  National User Facility through the Capability
  Computing Campaign (CCC) process. Run codes that
  can scale across a large fraction of the platform
  that are indicative of capability simulations.

15
Questions for you

• How are we doing at attracting the
  next-generation of weapons physicists, engineers
  and computer scientists?
  
• What are we missing in our march toward a
  predictive simulation capability?
  
• Can we do more in the area of scientific
  exchanges among the weapon laboratories and
  across the larger scientific community?
  
• Which NNSA applications can
• benefit from the ASC computing
• core competency?

Let me know your thoughts, either during this PI
meeting or via e-mail.