Transcript Introduction/Welcome by Fran Berman, Director of San

Cyberinfrastructure and
GEON
Dr. Fran Berman
Director, San Diego Supercomputer Center and NPACI
Professor and High Performance Computing Endowed
Chair, UCSD
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Science Today
Data
Management
and Mining
Physical Sciences
Digital
Libraries
Modeling and Simulation
GAMESS
Geosciences
Social
Sciences
Life
Sciences
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Technology as an enabler
• Today, modern applications
are enabled by
•
•
•
•
•
•
Networks
Visualization
Data storage
Remote instruments
Sensors
Supercomputers, servers,
laptops, handheld devices, etc.
• Integrated software
systems provide the
“glue”
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Science and Technology Bootstrapping
• Key components for
enabling new
discoveries include
• Expertise
• Software
• Services
• Resources
New
infrastructure
capabilities
enable
New
infrastructure
capabilities
motivate
New
application
goals
motivate
Application
goals
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Evolving MCell
• Need for larger-scale
MCell runs motivated
more efficient dataaware CS scheduling
CLIENT
motivate
APST
Middleware
enable
• CS middleware enabled
large enough runs that
computational steering
during MCell execution
could be explored
motivate
Bootstrapping larger-scale
simulations for MCell
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Cyberinfrastructure
• Cyberinfrastructure is the organized
aggregate of technologies enabling
access and coordination of information
technology resources to facilitate science,
engineering, and societal goals.
•
•
•
•
•
•
Data
Computation
Communication
Visualization
Scientific Instruments
Expertise, etc.
National Science Foundation’s
Cyberinfrastructure
NSF Blue Ribbon Panel
(Atkins) Report
provided a compelling
and comprehensive
vision of an integrated
Cyberinfrastructure
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Cyberinfrastructure can enable next-generation
applications
• Data Integration in the
Biosciences
Users
Software
to access
data
Software
to federate
data
• Data Integration in the
Geosciences
Disciplinary
Databases
Anatomy
What isotopic data exist for
region X?
How does it relate to local geology?
Organisms
Physiology
Data
Integration
Organs
Cell Biology
Cells
Complex
“multiple-worlds”
mediation
Proteomics
Organelles
Genomics
Biopolymers
Medicinal
Chemistry
Atoms
GeoGeologic
Chemical
Map
GeoPhysical
GeoChronologic
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Foliation
Map
Data Cyberinfrastructure from an IT Perspective
Applications: Medical informatics,
Biosciences, Ecoinformatics,…
Visualization
Data Mining, Simulation Modeling,
Analysis, Data Fusion
Knowledge-Based Integration
Advanced Query Processing
Grid Storage
Filesystems, Database Systems
High speed networking
Storage hardware
How do we represent data,
information and knowledge to
the user?
How do we detect trends and
relationships in data?
How do we obtain usable
information from data?
How do we collect, access
and organize data?
How do we configure computer
architectures to optimally support
data-oriented computing?
Networked Storage (SAN)
sensornets
How do we combine data, knowledge
and information management with
simulation and modeling?
instruments
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Community Projects:
Cyberinfrastructure’s First Responders
TeraGrid: “Top down”
Grid infrastructure
NEES: Earthquake
Engineering infrastructure
PRAGMA: Pacific Rim Grid
Middleware Consortium
BIRN: Biomedical Informatics
Research Network
NPACI Grid: “Bottom up”
Grid infrastructure
GEON: Geosciences
Grid infrastructure
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Solid Earth Science Cyberinfrastructure
CLIMATE CHANGE
LANDFORM RESPONSE
AND
AND
RESOURCE MANAGEMENT
MANAGEMENT
LEADS TO
LEADS TO
Cyberinfrastructure
Research
MANAGEMENT OF
DISASTER POTENTIAL
LEADS TO
LEADS TO
NITROGEN
CYCLE
CORE MANTLE
DYNAMICS
CRUSTAL DYNAMICS
CONTAIN
PALEO
ENVIRONMENTS
SUSTAINABILITY
OF
POPULATIONS
GEODYNAMICS
OTHERS
CARBON CYCLE
GLOBAL
GEOCHEMICAL
CYCLES
EARTHQUAKES
CONTAIN
GEOPHYSICAL
PROPERTIES
VOLCANOES
CONTAIN
MITIGATE GEOLOGIC
HAZARDS
RELATE TO
EVOLUTIONARY
RECORD
OTHERS
SOCIETAL
RELEVANCE
ABSOLUTE GEOLOGIC TIME
DERIVED
KNOWLEDGE
PROVIDE
COMPUTATIONAL, EXPERIMENTAL &
MODELING TOOLS
UTILIZE
FOSSIL RECORD
METAMORPHIC ROCKS
IGNEOUS ROCKS SEDIMENTARY ROCKS STRUCTURE OF ROCKS
*Based on: Summary of Global Overview ,National Research Council Report,1993 (by A.K.Sinha, Virginia Tech,2004)
SAN DIEGO SUPERCOMPUTER CENTER
PRIMARY
ROCK
RECORD
Slide courtesy of Krishna Sinha
University of California, San Diego
GEON: First Responder Cyberinfrastrucrure
• Pioneering GEON IT Components
•
Development of GEON Portal
Modeling, analysis, simulation, data
management
•
•
Implementation of a Geosciences-focused
•
Standards for information integration
•
•
•
Will link to TeraGrid
Data, metadata, ontology standards
Web/Grid service-based components
Community
•
•
Provide a unifying framework for meaningful
interpretation and understanding of Geoscience
data
Workshop to develop common standards for
schemas, ontology
•
•
Close collaboration between geoscientists
and technologists
Visualization
•
GEONgrid
•
Collaboration
•
Access
•
•
•
Visualization of 2D, 3D, 4D data;
information visualization to support
knowledge discovery
Augmented reality in the field
Framework for data
“transparency” on the Grid
•
•
•
•
•
Location/Name: data can be anywhere on
the Grid
Distribution: data can be distributed
across multiple locations
Replication multiple copies of data
Heterogeneity data can be in more than
one format—file or DBMS, points or
polygons, etc… (the difficult one)
Ownership & Costing should not have to
negotiate with each source individually
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Cyberinfrastructure Leverage and Synthesis
• GEON and BIRN will attempt to have common grid software stack
• GEON and SEEK will have common semantic integration services
• GEON is an application driver for OptIPuter
• Will field a common GIS and Viz/Synthesis center
• GEON is linked to Earthscope
• IRIS and UNAVCO are members of GEON. Both are primary data archives
for Earthscope
• USGS is a major partner of GEON
• Perhaps a GEON node(s) at USGS
• GEON is a partner of Chronos (chronostratigraphy)
• GEON is a partner of CUAHSI (hydrologic information systems)
• GEON will work with DOE EarthSystem Grid (ES Grid)
• ES Grid has deployed grid data services across 5 sites
• Ian Foster (Chicago), Don Middleton (NCAR), Ann Chernevak (ISI)
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Building Cyberinfrastructure
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Cyberinfrastructure Challenges
• Just building
Cyberinfrastructure makes
research questions out of
previously solved problems
•
•
•
Installation
Configuration
Accounting
• And brings added complexity
to existing problems
•
•
•
•
•
Performance analysis
Debugging
Scheduling
Security
Fault tolerance, etc.
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Cyberinfrastructure Challenges
•
Architecture: What is shared/private,
free/charged for, centralized/distributed, etc.
•
Administration: Who makes decisions?
Who enforces decisions?
•
•
•
•
Can I get an
account in Italy?
Australia?
Saudi Arabia?
•
What if I don’t
want to run your
application on my
resources?
•
What am I willing
to pay for?
•
Who makes
decisions? Who
enforces
decisions?
Sustainability: Who is responsible for
evolving infrastructure? Who is responsible
for supporting it?
Coordination: How do we link distinct
Cyberinfrastructure components, efforts,
software?
Scalability: How do we ensure stability as
Cyberinfrastructure software and
communities grow?
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
New Challenges brought on by Broad Usage
• Social, economic, political
policies for
Cyberinfrastructure are
largely uncharted territory
Cyberinfrastructure
environments will need to
be robust, interactive,
dynamic, and userfriendly.
Cyberinfrastructure user
environments may look
more like role-playing
games than today’s
scientific portals.
Aggregate user
behavior must not
destabilize the
system
Cyberinfrastructurewide policies will be
key to avoiding
resource instability,
thrashing, and
“blackouts”
Cyberinfrastructure
“economics”:
Balancing cycles,
bandwidth and
storage presents a
“multiple currencies,
multiple users”
problem
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Next Generation Challenges:
Cyberinfrastructure Evolution
• What should our aggregate Grid/Cyberinfrastructure/integrated
technology platforms look like in 10 years?
•
What kind of research and development should we be encouraging now?
•
What do we need to preserve now that will be useful in the future?
•
What are the building blocks? What are the goals? What are the metrics of
success? How should we evaluate where we’re at?
•
Lessons learned from the past:
• Don’t oversell the vision.
• Don’t underestimate what it takes to get there.
• It’s not successful infrastructure until the users say it is.
Research
Development
Deployed
Infrastructure
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
GEON is a point of the Spear
• Key challenges and
opportunities critical for
successful Cyberinfrastructure
being addressed first by
GEON.
• GEON’s success is important
for Cyberinfrastructure’s
success.
GEON is a driving project for SDSC
and we are committed to enabling its success.
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego
Welcome to the GEON All
Hands Meeting and SDSC
SAN DIEGO SUPERCOMPUTER CENTER
University of California, San Diego