CyberInfrastructure ( CI ): Whence? Thomas J. Greene, Ph.D. Sr. Program Dir. for CISE-ANIR, National...
-
Upload
tatiana-solloway -
Category
Documents
-
view
219 -
download
1
Transcript of CyberInfrastructure ( CI ): Whence? Thomas J. Greene, Ph.D. Sr. Program Dir. for CISE-ANIR, National...
CyberInfrastructure (CI ): Whence?
Thomas J. Greene, Ph.D.Sr. Program Dir. for CISE-ANIR,
National Science Foundation
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 2
Outline
1. A vision of information/knowledge progress
2. The state of NSF CISE vision of CI
3. Some thoughts on a strategy
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 3
This big Vision is :Cyberinfrastructure
An old idea of information processing using machines made from :
• INPUT/OUTPUT devices• PROCESSORS
• MEMORY• CONNECTIONS
Becomes a new idea through:• Very BIG Sizes and • GLOBAL distances
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 4
Outline
1. A vision of information/knowledge progress
2. The state of NSF CISE vision of CI
3. Some thoughts on a strategy
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 5
Evolution of the Computational Infrastructure
Supercomputer Centers
PACI
Terascale
1985 1990 1995 2000 2005 2010
| | | | | |
NPACI and Alliance
SDSC, NCSA, PSC, CTC
TCS, DTF, ETF
Cyberinfrastructure
Prior Computing Investments
NSF Networking
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 6
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
http://www.communitytechnology.org/nsf_ci_report/
Setting the Stage
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 7
Cyberinfrastructure Promise
• Ubiquitous, digital knowledge environments that are both interactive and functionally complete…………
• revolutionize the processes of discovery, learning and innovation across the science and engineering frontier.
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 8
Characteristics of Cyberinfrastructure
• Community-Driven• Distributed Collaboration• Virtual Organization• Multidisciplinary in scale and scope• International in scale and scope• Interoperable• Supporting Data- and Compute-Intensive Applications• High end to desktop• Distributed• Heterogenuous• Complex• Reusable
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 9
Cyberinfrastructure Early Adopters
• Network for Earthquake Engineering Simulation (NEES)
• National Ecological Observatory Network (NEON)
• Biomedical Informatics Research Network (BIRN)
• Extensible Terascale Facility (ETF)
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 10
CalTech Argonne
Terascale Extensions Program
SDSC
LosAngeles
Chicago
NCSA PSC
Existing ETF Partners
Hubs
New Partners
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 11
FY2003 ETF Enhancements
• Solicitation to attract new resource partners (due June 9th)
• Funding connection and integration• Resources may include:
– Archival repositories– Digital libraries– Computational resources– Sensor networks
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 12
Computation
Communication Information
Collaboration
Community
Culture
Cyberinfrastructure
KNOWLEDGE
On The Path to Knowledge
DATA
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 13
Hardware
Integrated CI System meeting the needs of a community
of communities
Grid Services & Middleware
DevelopmentTools & Libraries
Applications• Environmental Science• High Energy Physics• Proteomics/Genomics• …
Domain-specific
Cybertools (software)
Domain-specific
Cybertools (software)
Shared Cybertools (software)
Shared Cybertools (software)
Distributed Resources
(computation, communication, storage, etc.)
Distributed Resources
(computation, communication, storage, etc.)
Ed
uca
tion a
nd
Tra
inin
g
Dis
covery
& In
novati
on
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 14
Shared Cybertools(Middleware Tools and Services)
Basic Services
Security, Scheduling, Data Services, Database Services, User Services, Application Management Services, Autonomy and Monitoring Services, Information Services, Composition Service, Messaging Service
Application Level ServicesPeople Collaboration, Resource Collaboration,
Decision-Making Services,Knowledge Discovery
Services, Workflow Services, Universal Access
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 15
Challenging Context
• Institutional & Infrastructural Ecology– Technological change more rapid than institutional change
• Broadening Participation• Community-Building – Role of early adopting
communities as drivers/partners ?• Seamless Integration of New and Old
– Balancing upgrades of existing and creation of new resources– Legacy data/models
• Providing sustainable support
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 16
Plan of Action
• Focused, cross-cutting attention on cyberinfrastructure – not business as usual
• Internal NSF planning now underway - active discussion on specific cyberinfrastructure issues
• Community building - broad consultations with scientific communities will intensify in coming months
• Summer 2003 workshops and town hall meetings – management models
• NSF FY05 budget planning for cyberinfrastructure beginning shortly
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 17
Outline
1. A vision of information/knowledge progress
2. The state of NSF CISE vision of CI
3. Some thoughts on a strategy… (a modest proposal ?)
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 18
Why collaborate ?
• (1) Altruism
• (2) Self- Interest – For this project choose (2) because busy
communities often respond easily to Self-interest (especially with visible Leveraged outcomes for the community agenda)
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 19
A CS Strategy for CI building ?
• PICK some existing global big science projects
• Engage them in a revolution using projects building/using cyberinfrastructure components
• Agree to Find “de facto standards” for mutual benefit across cultural divides
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 20
Leverage at the Interface(Equal Partnership Collaborations)
• SCIENCE APPLICATIONS
• COMPUTER TOOLS
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 21
A Science Strategy for CI building ?
• PICK some existing computer science projects
• Engage them in a revolution using projects building/using cyberinfrastructure components
• Agree to Find “de facto standards” for mutual benefit across cultural divides
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 22
CI Is a Large SYSTEM so build it as most very large systems are
built-- – Bottom up --- micro– Top down --- macro– Middle out – engineering choices for standard
interfaces (Standards solutions etc.)
• --• USE Cyber infrastructure rapid proto demos• And keep them alive after the demo …
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 23
Cyberinfrastructure vision involves a Large COMPLEX system – (with Synergy?)
• top down
• bottom up
• or middle out
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 24
Bad News
• Crossing cultures to enable collaborations is not a well understood process.
• New tools for this are needed in computers and communications.
• Agreements to enabling standards requires people compromising to agree to further the big agenda – “big thinkers”.
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 25
Good News - a Win-win Situation Exists
• Essential changes in information access are happening now, (not just a little faster, bigger but - much faster, bigger, wider!).
• Solving very hard problems faster and better by the collaborations “idea” by some frontiersmen is already agreed to .
• Committed problem solvers will “climb mountains “ to solve their problem agenda and will even work with people outside their culture – sometimes.
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 26
Some Principles for a Global Cyber Infrastructure and for E-science Rapid Prototyping
Some proposed principles under current discussion 0] The cost and complexity of 21st Century Science requires the creation
of advanced and coherent global Infostructure (information infrastructure).
1] The construction of a coherent Global Infostructure for Science
requires definition and drivers from Global Applications (that will also communicate with each other)
2] Further, forefront Information Technology must be incorporated into
this Global lnfostructure for the Applications to reach their full potential for changing the way science is done
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 27
Some CI Building Tactics
• Use common agendas, not common check books.( i.e. Keep money with owning agencies as much as possible.)
• Full credit to all
• Give Endorsements across cultures whenever useful.
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 29
Acknowlegement
The members of the NSF CyberInfrastructure Working Group
(CIWG)
Dr. Deborah Crawford,Chair
TJG -- 21st NORDUnet conference --Iceland - 24/AUG/2003 30A global vision of Ubiquitous information at Light-speed– Cyberinfrastucture ( Grids, E-science) =
globe