Directors View

1
Directors View

• NPACI All-Hands Meeting
• March 6-9, 2002
• Francine Berman
• Director, NPACI and SDSC
• berman_at_sdsc.edu

2
Welcome, Everyone!
Your Logo Here
PSC
3
The NSF PACI Program

• PACI program is developing the human, software,
  data, networking and computational infrastructure
  to address critical science and technology needs
• Focus is on the development of large-scale
  infrastructure
• Broad mission includes development of
  infra-structure to promote scientific advances,
  community access, education
• PACI program instrumental in development of
  data-intensive computing, Grid computing

4
Science that Makes a Difference

• NPACI human, software and hardware infrastructure
  is key to enabling the next generation of
  advances in technology, science, and engineering.

Deep Impact
Broad Impact
Individual Impact
5
Deep Impact

• RAMS simulations link climate models to land
  cover and biodiversity models
• Eastman, Pielke (Colorado State) with NPACI ESS
  partners LTER, U Kansas, SDSC
• Microtubule properties two orders of magnitude
  increase in atoms simulated
• McCammon, et al., UCSD
• Exploration and visualization of ensembles of oil
  reservoir simulations
• Alpha project with the Center for Subsurface
  Modeling at U Texas, PTE partners at Ohio State
  U, U Maryland

6
Broad-based Impact
Selected NPACI Software Deployment
NPACI HPC Resources
800 new user logins created in the last year
1200 users used allocated time in the last
year 7,200 total publications theses citing
NPACI/SDSC resources 14.6M hits to NPACI and
SDSC web servers on average per month 9.5M
service units allocated in the last year
NPACI Archives
470 Terabytes of Data stored at Resource
Partners
DataCutter
Almost every Grid initiative is using
NPACI-supported technologies includingBIRN,
NVO, SCEC, GriPhyN, iVDGL, MCell, TeraGrid
7
Individual Impact

• Tapia Celebration of Diversity in Computing, Oct
  18-20, Houston, TX
• Ed Center launches Computational Science Resource
  Community
• Shodor Foundation awarded 2.7 million from NSF
  to extend National Computational Science
  Institute
• CRA-W starts eighth year of its Distributed
  Mentoring Program

8
Driving a National Information Infrastructure
August 9, 2001 NSF Awarded 53,000,000 to
SDSC/NPACI and NCSA/Alliance for TeraGrid

• TeraGrid will provide in aggregate
• 13.6 trillion calculations per second
• Over 600 trillion bytes of immediately accessible
  data
• 40 gigabit per second network speed
• TeraGrid will provide a new paradigm for
  data-oriented computing
• Critical for disaster response, genomics,
  environmental modeling,

9

• PIs Berman, Foster, Messina, Reed, Stevens
• Original Sites SDSC/UCSD, Caltech, NCSA/UIUC,
  ANL
• Partners IBM, Intel, Qwest, Sun, Myricom,
  Oracle and others
• Cool Things about the TeraGrid
• Big data, simulation, modeling
• Grid computing, Globus, portals, middleware
• Clusters, Linux
• Production facility, usability key
• Scalability is fundamental

• TeraGrid Software Environment
• Linux
• Basic and Core Globus Services
• Advanced Services
• Data Services
• Over .6 Petabytes of on-line disk will provide
  ultimate environment for data-oriented
  computation
• Linux environment provides more direct path from
  development on lab cluster to performance on
  high-end platform

10
Looking Ahead

• Science needs are evolving
• Data motivates fundamental infrastructure
  challenges for the next decade
• Technology is changing
• Array of computer, network, storage, instruments
  and devices increasingly heterogeneous
• Increasing need for many components to coordinate
  
• Globalization of computation
• NSFs vision is evolving
• Cyberinfrastructure ? National Information
  Infrastructure
• Blue Ribbon Panel recommendations will help set
  the stage
• Terascale computing will provide the basis

11
What has NSF been doing this year?

• Blue Ribbon Panel on Cyberinfrastructure
• Evaluate the performance of the PACI Program in
  meeting the needs of the scientific research and
  engineering community.
• Recommend new areas of emphasis for the NSF CISE
  Directorate that will respond to the future needs
  of this community.
• Recommend an implementation plan to enact any
  changes anticipated in the recommendations for
  new areas of emphasis.
• Will most likely provide recommendations in
  April/May
• Terascale Computing System initiative
• TCS 2 Solicitation resulted in successful
  NPACI/Alliance TeraGrid proposal
• TCS 3 Solicitation coming soon

12
Blue Ribbon Panel

• Broad themes from BRP Testimony (1)
• Information Infrastructure is a fundamental tool
  for science.
• Grid Computing is essential.
• The U.S. must build a persistent, evolutionary,
  grid-based National Information Infrastructure to
  stay competitive and maintain its leadership
  role.
• Dealing with immense amounts of data is the
  critical problem of the next decade.

13
Blue Ribbon Panel

• Broad themes from BRP Testimony (2)
• The PACI program and the TeraGrid project can
  provide the core of a National Information
  Infrastructure.
• Substantive coordination, software and human
  infrastructure is critical for the development of
  a National Information Infrastructure usable and
  relevant to a broad spectrum of science and
  society.

14
NSF Transitions

• NSF CISE AD
• Peter Freeman, Dean of Computing, GA Tech coming
  on
• Ruzena Bajcsy now Director of the Center for
  Information Technology Research in the Interest
  of Society at UC Berkeley
• NSF ACIR Division Director
• Rich Hirsh, Acting Director
• Bob Borchers now Chief Technology Officer, Maui
  High Performance Computing Center

15
NPACI Transitions

• LT
• Paul Messina, retiring Chief Architect
• Carl Kesselman joining as Chief Software
  Architect
• Peter Arzberger joined UCSD administration as
  Director, Life Sciences Initiatives
• Richard Moore joined NPACI as Executive Director
• Peter Taylor leaving for University of Warwick,
  Professor of Chemistry and Chief Scientist at the
  Centre for Scientific Computing (Summer, 2002)

16
NPACI Transitions

• EC
• Metasystems/Grid Computing Lead changed from
  Andrew Grimshaw to Rich Wolski and Carl Kesselman
  to Rich Wolski
• ESS Lead changed from Jim Beach (elected Chairman
  of the Board of Trustees of BIOSIS) and Cherri
  Pancake to Cherri Pancake
• EOT-PACI
• Rozeanne Steckler joined EOT-PACI LT
• Ann Redelfs leaving as co-thrust lead for dream
  job in Cornell Laboratory of Ornithology
• Russ Altman stepping down as Molecular Science
  Thrust Lead (new Director of the Stanford Center
  for Biomedical Computation)
• Jim Pool returning as Member at Large

17
2002 NPACI Executive Committee
?
Rich Wolski, UC Santa Barbara Metasystems/Grid
Computing Joel Saltz, Ohio State Programming
Tools and Environments Reagan Moore,
SDSCData-Intensive Computing Arthur Olson,
TSRIInteraction Environments William Martin, U
MichiganResource Representative Jim Pool,
Caltech Member at Large
TBA, Molecular Science Mark Ellisman,
UCSDNeuroscience Cherri Pancake, Oregon State
Earth Systems Science Tinsley Oden, U
TexasEngineering Gwen Jacobs, Montana
StateMember at Large Aron Kuppermann,
Caltech User Representative
18
2002 NPACI Executive Committee and Leadership
Team

• Susan Graham, UC Berkeley Chief Computer
  Scientistgraham_at_cs.berkeley.edu
• Peter Taylor, SDSCChief Applications
  Scientisttaylor_at_sdsc.edu
• Wayne Pfeiffer, SDSCResources pfeiffer_at_sdsc.edu
• Greg Moses, U WisconsinEducation, Outreach, and
  Training Thrust Leadmoses_at_engr.wisc.edu

Francine Berman, SDSCDirectorberman_at_sdsc.edu Ric
hard Moore, SDSC Executive Directorrlm_at_sdsc.edu C
arl Kesselman, Chief Software Architectcarl_at_isi.e
du
19
Last Years AHM Homework Assignment

• At last years All Hands Meeting, we sought to
  evaluate the NPACI vision, strategy and
  organizational structure.
• We focused on the following questions
• How are things working?
• How do the NPACI activities fit together as a
  whole?
• How do we measure our impact?

20
What we did

• EC and LT spent most of the year in a frank and
  thoughtful evaluation of NPACIs priorities and
  organizational structure
• A set of strategic priorities for the partnership
  were developed (next slide)
• Considered many organizational alternatives from
  staying with the current system to radical
  restructuring
• Punchline The basic structure is good but we
  want to
• Increase internal and external communication
• Increase accountability
• Coordinate the NPACI infrastructure portfolio
  more effectively

21
NPACI Strategic Priorities

• Define future directions for high-end scientific
  computing and field the national infrastructure
  to enable them.
• Expand and diversify the community of users of
  NPACI resources.
• Maximize the impacts of NPACI technologies on the
  research community.
• Increase the understanding of NPACI technology
  for scientific computing and scientific data
  environments for a broad community.

22
NPACI Organizational Improvements

• Ad hoc committees to help NPACI work more
  effectively
• Strategic Projects defining, guiding and
  coordinating thrust activities in both the
  applications and technology domains
• Resources and Users defining, guiding, and
  coordinating the portfolio of production-level
  hardware, software, and services
• Outreach and Coordination defining and guiding
  EOT, coordinating internal communication,
  international affiliates, AHM, SCxy, new
  communities
• Committees will operate for next 6 months, then
  be evaluated by EC
• Increased focus on review and strategic planning
• Continue annual review of all projects
• Strategic priorities will be used to help
  evaluate projects

23
What does the Future Hold?

• In the next decade, our National Information
  Infrastructure will need to support unprecedented
  
• Diversity
• Globalization
• Integration
• Data management, analysis and knowledge synthesis
• Scale
• Use

SensorsExperiments, Instrumentation
24
Fundamental IT Challenges for the Next Decade

• Data collection to knowledge synthesis, data
  mining
• Data ? Information ? Knowledge
• How do we determine usable information from
  massive amounts of data?
• Integrated technologies
• Sensors, computation, data management
• How do we develop end-to-end applications which
  integrate new technologies?

25
Fundamental IT Challenges for the Next Decade

• Universal Access
• Location-independence, ubiquitous access to
  computing, data, resources
• How do we develop applications which can be
  initiated from anywhere at anytime?
• Global-scale collaboration
• Collaboration tools, access, interoperability
• How do we communicate, work, and recreate with
  people all over the world?

26
Infrastructure for Today and Tomorrow

• Information infrastructure required for the next
  generation of scienceand technology results
• Unprecedented partnership required in order to
  meet the challenges of the next decade
• PACI program and TeraGrid provide the core of a
  National Information Infrastructure

27
Welcome to AHM 2002

• The future National Information Infrastructure
  starts here.
• Participate actively! This is your partnership.
• Enhance your own collaborations
• Help us build a stronger and more successful
  partnership
• Tell us what you think.
• Talk with EC members and other partners
• Please give your feedback and return surveys
• Thanks in advance for providing input to the
  NPACI Communications group
• Thanks to IBM for funding the AHM Reception!

28
Thanks

• Paul Messina
• Peter Arzberger
• Peter Taylor