Cyberinfrastructure (CI) "It is easier to invent the future than to predict it" - Alan Kay Clifford A. Jacobs National Science Foundation Division of Atmospheric Sciences Head, UCAR and Lower Atmospheric Facilities Oversight Section 4201 Wilson Blvd, Suite 775 Arlington, VA

Outline  Background  Revolutionizing Science and Engineering through cyberinfrastructure  Planning for Cyberinfrastructure  Conclusions

Background An incomplete of historical events that set the stage for CI

Panel on Large Scale Computing in Science and Engineering (Lax report, 1982)  Increase access to regularly upgraded supercomputing facilities via high bandwidth networks.  Increase research in computational mathematics, software, and algorithms.  Train people in scientific computing.  Invest in research on new supercomputer systems.

NSF Blue Ribbon Panel on High Performance Computing (1993) A. CENTRAL GOAL FOR NSF HPC POLICY –VISION OF THE HPC PYRAMID B. RECOMMENDATIONS TO IMPLEMENT THESE GOALS C. THE NSF HPC CENTERS –ALLOCATION OF CENTER HPC RESOURCES TO INVESTIGATORS –EDUCATION AND TRAINING D. NSF AND THE NATIONAL HPC EFFORT; RELATIONSHIPS WITH THE STATES

NSF Blue Ribbon Panel on High Performance Computing (1993) Challenges I. How can NSF remove existing barriers to the rapid evolution of HPC, making it truly usable by all the nation’s scientists and engineers? II. How can NSF provide scalable access to a pyramid of computing resources?  Balance computing environments and access III. How can NSF encourage the continued broadening of the base of participation in HPC? IV. How can NSF best create the intellectual and management leadership for the future of HPC in the U.S.? RECOMMENDATIONS 1. Balanced pyramids of computing environment 2. Essential research investments to remove the obstacles to realizing the technology of the pyramid and its effective use. 3. Institutional structure for delivery of HPC capabilities 4. NSF’s role at the national level and its relationship with the States in HPC

Hayes Report Vision (1995)  Provide access to high-end computing infrastructure for the academic scientific and engineering community;  Partner with universities, states, and industry to facilitate and enhance that access;  Support the effective use of such infrastructure through training, consulting, and related support services;  Be a vigorous early user of experimental and emerging high performance technologies that offer high potential for advancing computational science and engineering;  Facilitate the development of the intellectual capital required to maintain world leadership.

Evolution of the Computational Infrastructure Supercomputer Centers PACI | | | | | | NPACI and Alliance SDSC, NCSA, PSC, CTC Prior Computing Investments NSF Networking From NSF/CISE Directorate Lax1982 Branscomb1993Hayes1995

Enabling and Motivating a CI Initiative ASC PACI’s Pittsburgh TSC Distributed Terascale Facility Some ITR Projects Digital Library Initiatives Networking Initiatives Middleware Initiatives Other CISE Research Cyber- Infrastructure Initiative Initiatives in non-CISE Directorates NSB Research Infrastructure Review Initiatives in DOE, NIH, DOD, NASA, … International Initiatives: UK e-science, Earth Simulator, EU Grid & 6th Framework Scientific Data Collection/Curation Collaboratories

NSB asked in 2001: What is NSF’s Strategy to provide CI to the S&E Community? Evolution of the Computational Infrastructure Supercomputer Centers PACI | | | | | | NPACI and Alliance SDSC, NCSA, PSC, CTC Terascale TCS, DTF, ETF High-End Cyberinfrastructure Prior Computing Investments NSF Networking From NSF/CISE Directorate HPCC Grand Challenge PITAC ITR NSB Infrastructure ReportAktins2003

Blue Ribbon Panel convened by the Assistant Director for Computer and Information Science and Engineering (CISE) of the National Science Foundation (NSF) – –Daniel E. Atkins, Chair   University of Michigan – –Kelvin K. Droegemeier   University of Oklahoma – –Stuart I. Feldman   IBM – –Hector Garcia-Molina   Stanford University – –Michael L. Klein   University of Pennsylvania – –David G. Messerschmitt   University of California at Berkeley – –Paul Messina   California Institute of Technology – –Jeremiah P. Ostriker   Princeton University – –Margaret H. Wright   New York University  1) evaluate current major investments in cyberinfrastructure, most especially the Partnerships for Advanced Computational Infrastructure (PACI)  2) recommend new areas of emphasis relevant to cyberinfrastructure; and  3) propose an implementation plan for pursuing these new areas of emphasis. PanelCharge

Input to Panel  62 presentations at invitational public testimony sessions  700 responses to a community-wide survey  review of dozens of prior relevant reports; scores of unsolicited s and phone calls  250 pages of written critique from 60 reviewers of an early draft of this report  hundreds of hours of deliberation and discussion between Panel members  The members of the Panel have backgrounds in areas widely relevant to creating, managing, and using advanced cyberinfrastructure. courtesy of D. Atkins

LIGO ATLAS and CMS NVO and ALMA The number of nation-scale projects is growing rapidly! Climate Change Cyberinfrastructure Enabled Science courtesy of D. Atkins

Revolutionizing Science and Engineering through Cyberinfrastructure  NSF has an opportunity to provide leadership for the Nation in an initiative to revolutionize science and engineering research capitalizing on cyberinfrastructure opportunities. –A nascent revolution has begun. Demand is here and growing. The time is now (opportunities & opportunity costs.) –Many prior investments (projects, initiatives, centers) are a key resource to build upon. –Now need sanction, leadership and empowerment through significant new funding and effective coordination. –Need very broad (synergistic) participation by many communities with complementary needs and expertise. –Need appropriate leadership and management structure. –Need incremental funding of $1B/year.

Cyberinfrastructure Vision  provide an integrated system of hardware and software resources and services that...  enable scientists and engineers to explore important research and education opportunities… that otherwise would not be possible

Cyberinfrastructure (CI) Overarching recommendation “NSF should establish and lead a large- scale, interagency, and internationally coordinated Advanced Cyberinfrastructure Program (ACP) to create, deploy, and apply cyberinfrastructure in ways that radically empower all scientific and engineering research and allied education” § Cyberinfrastructure is "essential, not optional, to the aspirations of research communities." Cyberinfrastructure is "essential, not optional, to the aspirations of research communities."

Opportunity  Enable more ubiquitous, comprehensive knowledge environments that become functionally complete for specific research communities in terms of –People –Data –Information tools –Instruments  “Such environments enables teams to share and collaborate over time and over geographic, organizational and disciplinary distances.”  Can serve individuals, teams, organizations to revolutionize –What they can do –How they do it –Who participates

Shared Opportunity and Responsibility  All NSF communities  Multi-agency  Industry  International

Cyberinfrastructure consists of …  Computational engines  Mass storage  Networking  Digital libraries/data bases  Sensors/effectors  Software  Services  All integrated to permit the effective and efficient building of applications

Integrated CI Services to Enable New Knowledge Environments for Research and Education High performance computation services Data, information, knowledge management services Observation, measurement, fabrication services Interfaces, visualization services Collaboration services Networking, Operating Systems, Middleware Base Technology: computation, storage, communication Customization for discipline- and project specific applications Community-Specific Environments for Research and Education (collaboratory, co-laboratory, grid community, e-science, virtual community) Cyberinfrastructure

Dimensions of CI Research courtesy of D. Atkins

Interactivity: Converging Streams of Activity GRIDS (broadly defined) E-science CI-enabled Science & Engineering Research & Education Science-driven pilots (not using above labels) ITFRU Scholarly communication in the digital age Home Land Security courtesy of D. Atkins

Computational Diversity Capability not just capacity: technology, policy, tools. Still need some center-based leading- edge super computers. On-demand supercomputing,not just batch. courtesy of D. Atkins

Pervasiveness of CI-enabled Knowledge-Based Activities Opportunities and challenges for multi-use. Opportunities for leverage and coalition building to obtain the resources to really do it. courtesy of D. Atkins

Advanced Cyberinfrastructure Program (ACP): Need highly coordinated, persistent, major investment in…  Research and development (CI as object of R&D) –Base technology –CI components & systems –Science-driven pilots  Operational services –Distributed but connected (Grid) –Exploit commonality, interoperability –Advanced, leading-edge but… –Robust, predictable, responsive, persistent  Domain science communities (CI in service of R&D) –Specific application of CI to revolutionizing research (pilot -> operational) –Required not optional. New things, new ways. –New things, new ways. Empowerment, training, retraining. X-informatics.  Education and broader engagement –Multi-use: education, public science literacy –Equity of access –Pilots of broader application: ITFRU, industry, workforce & economic development courtesy of D. Atkins

Incremental budget estimates based on  current and previous NSF activities  testimonies  other agencies’ programs in related areas  activities in other countries  explicit input from community on first draft courtesy of D. Atkins

Major components of ACP Total Annual Budget $1020 (Annual expenditure in millions of dollars) Provision of operational operational CI * CI * Application of CI to advance S&E research research Information and data support Fundamental research to advance CI Computationalcenters Data repositories Digital Libraries Networking & connections Application service centers $375 $185 $30 $60 $10 $660 $200 $200 $60 * Includes $75M/Year for computational centers

NOT Business as Usual for NSF nor It’s Usual Research Community  Scale  Duration, Persistence  Complexity (technical, managerial)  Richness and Diversity of Partnerships  Gaps, Chasms  Approach: Pasteur Quadrant model of knowledge creation and use.  (Also more persistence and scale in provisioning of CI for the research community.) courtesy of D. Atkins

National Petascale Systems National Petascale Systems Ubiquitous Sensor/actuator Networks Ubiquitous Sensor/actuator Networks Laboratory Terascale Systems Laboratory Terascale Systems Ubiquitous Infosphere Collaboratories Responsive Environments Responsive Environments Terabit Networks Contextual Awareness Contextual Awareness Smart Objects Smart Objects Building Out Building Up Science, Policy and Education Petabyte Archives Petabyte Archives Futures: The Computing Continuum courtesy of D. Atkins

Planning for Cyberinfrastructure

Cyberinfrastructure Planning: Steps/Timeline  Broad, informal consultations with community - continuous  Internal NSF planning underway since January  Workshops to seek community inputs on management structures  NSF FY05 and beyond budget planning for CI - underway  ETF Management and Operation Approach to be developed by end of this summer  Broad NSF CI Roadmap, v1 - early 2004 From presentation to NSB Aug. 2003

Recent Planning Activities  Management & Models Meetings (4 total)  Budget planning (across NSF)  Cyberinfrastructure Working Group (CIWG) active across directorates at program level  Strategic planning underway (CISE leading)  Programmatic plans expected by mid- October  Proposed CISE reorganization From presentation to NSB Aug. 2003

Cyberinfrastructure Planning: Strategies  Exploit the opportunities open to us to revolutionize S&E  Maintain technological leadership in computation, information, communications resources  Balance upgrades and new resources  Develop structures to permit significant growth in funding  Aggressively seek expanded CI funding From presentation to NSB Aug. 2003

Other Updates  House Committee on Science testimony –Committed to key recommendations of Atkins Report –NSF will continue commitment to supercomputing –Supercomputing is essential part of cyberinfrastructure vision –Grid investments underscore importance of integration  Terascale Extensions competition From presentation to NSB Aug. 2003

Other Updates (continued)  High-End Computing Revitalization Task Force (HEC-RTF) and cooperation with other agencies  Industry involvement (Council on Competitiveness study) From presentation to NSB Aug. 2003

Conclusions  CI is not new! –Atmospheric Sciences has been engaged building CI for more than five decades  What is new? –A holistic view of CI –The enormity of opportunities and challenges that CI offers –The need to lead and administer CI with coordinated leadership  CI will enrich Atmospheric Sciences

Atkins Report: Clifford A. Jacobs National Science Foundation Division of Atmospheric Sciences Head, UCAR and Lower Atmospheric Facilities Oversight Section 4201 Wilson Blvd, Suite 775 Arlington, VA

Pasteur’s Quadrant Research Model (from Donald Stokes) Focus on New Knowledge Creation? Focus on Application? No Yes Edison Bohr Pasteur courtesy of D. Atkins

ACP Report:

Layered Architecture of the ACP Applications of information technology to science and engineering research Conduct of science and engineering research Cyberinfrastructure supporting applications Core technologies incorporated into cyberinfrastructure Cyberinfrastructure Cyberinfrastructure