Transcript Slide 1
The
eXtremeDigital (XD)
Program
Barry I. Schneider
Office of Cyberinfrastructure
National Science Foundation
email:[email protected]
eXtreme
complex
XD Solicitation/XD Program
Digital Resources for Science and Engineering (NSF 08-571) – eXtremely
High-Performance Computing and Storage Services
• aka Track 2 awardees and others become service providers (SP) for XD
High-Performance Remote Visualization and Data Analysis Services
• 2 awards; 5 years; $3M/year
• proposals due November 4, 2008 – Two awards funded in 2009 – RDAV (NICS) & Longhorn (TACC)
Integrating Services (5 years, $26M/year)
• Coordination and Management Service (CMS)
– 5 years; $12M/year
• Technology Audit and Insertion Service (TAIS)
– 5 years; $3M/year – Two awards funded in 2010 – Buffalo (TAS) and UIUC (TIS)
• Advanced User Support Service (AUSS)
– 5 years; $8M/year
• Training, Education and Outreach Service (TEOS)
– 5 years, $3M/year
two phase proposal process for IS
• pre-proposals November 4, 2008
• final proposals due June 15, 2009 – Final project called XSEDE funded July, 2011
for five years with a possible additional five year extension
XD at a Glance
Technical Audit Service develops lightweight tools (XDMoD) to enable
stakeholders to obtain information on the systems, performance, usage and
other characteristics of the XD resources – extremely valuable for data
collection – Aid users, system administrators and NSF Program Director
Technical Insertion Service evaluates software and middleware and makes
recommendations for injection into project
XSEDE provides the integrating services for the project: Coordination and
management, advanced user services and training, education and outreach
XSEDE is a comprehensive set of advanced heterogeneous high-end
digital services, integrated into a general-purpose infrastructure.
XSEDE is about increased user productivity, not just high end resources
Increased productivity leads to more science at all levels
Increased productivity is sometimes the difference between a feasible
project and an impractical one
Thanks to the XSEDE folks for permission to use
material from slides used at the US-EU Summer
School in computational science held at Lake Tahoe
in early August
4
The Vision
Enhance the productivity of scientists and
engineers by providing them with new and
innovative capabilities
Facilitate scientific discovery while enabling
transformational science/engineering and
innovative educational programs
Develop a more sustainable approach
to the national CI
XSEDE supports a breadth of research
From direct contact with user community as part of
requirements collections
Earthquake Science and Civil
Engineering
Brain science
Analysis of large cosmological
simulations
Molecular Dynamics
Nanotechnology
DNA sequencing
Plant Science
Storm modeling
Computational Molecular
Sciences
Epidemiology
Neutron Science
Particle Physics
Economic analysis of phone
network patterns
International Collaboration in
Cosmology and Plasma Physics
Sampling of much larger set. Many examples are new to TeraGrid/HPC.
Range from petascale to disjoint HTC, many are data driven. XSEDE will
support thousands of projects.
XSEDE’s Distinguishing Characteristics
Foundation for a national CI ecosystem – in sync with the
NSF CIF21 vision
A comprehensive suite of advanced digital services
Federated with other high-end facilities, other
providers of digital services here and abroad
and importantly campus-based resources
Unprecedented integration of diverse digital resources
Developing an innovative, open architecture making
possible the continuous addition of new technology
capabilities and services – Achieve Seamless
integration of resources at
all levels
XSEDE: An Infrastructure Designed for
Innovation & Evolution
An environment in which all resources, data and services relevant
to a researcher can be embedded and shared
Campus bridging creating a single virtual system with interactive data
transfer and resource sharing capabilities
• “make my data accessible everywhere I want to be”
Coordinated archival approach to ensure persistence of important
datasets beyond the lifetime of particular service providers – Still needs
fleshing out
An underlying infrastructure to support this
Open architecture with judicious use of standards designed to evolve in a
non-disruptive way
Interoperability of XSEDE with other Cis – OSG, PRACE, DEISA
NAREGI
XSEDE’s Distinguishing Characteristics Architecture
XSEDE
is designed for innovation & evolution
There is an architecture defined, but it is still evolving as a consequence of
the review process
• based on set of design principles
• rooted in the judicious use of standards and best practices
• clearly defined transition plan from TeraGrid to XSEDE
Professional
systems engineering approach
Responds to evolving needs of existing, emerging, and new communities
• incremental development/deployment model
New requirements gathering processes
• ticket mining, focus groups, usability panels, shoulder surfing
Ensure robustness and security while incorporating new and improved
technologies and services
Process control, quality assurance, baseline management, stakeholder
involvement
How XSEDE describes the architecture
A
set of “views” examining the elephant from the perspectives of different
stakeholders
Not (only) immensely detailed documentation!
Different
stakeholders require different views, e.g.,
Service provider
System administrator
Power user
Occasional user
Gateway developer
Tell
us what views you think
are important
–Security officer
–NSF program manager
–Campus CIO
–Trainer
–...
XSEDE’s Distinguishing Characteristics Governance
World-class
leadership from CI centers with deep experience: partnership
led by NCSA, NICS, PSC, TACC and SDSC
PI:
John Towns,
Co-PIs: Jay Boisseau,
Patricia Kovatch,
Ralph Roskies,
Nancy Wilkins-Diehr,
NCSA/Univ of Illinois
TACC/Univ of Texas Austin
NICS/Univ of Tenn-Knoxville
PSC/CMU
SDSC/UC-San Diego
Partners
who strongly complement these CI centers with expertise in
science, engineering, technology and education
Univ of Virginia
SURA
Indiana Univ
Univ of Chicago
Berkeley
Shodor
Ohio Supercomputing Center
Cornell
Purdue
Rice
NCAR
Jülich Supercomputing Centre
Engaging stakeholders
Collection
of stakeholder needs:
surveys, ticket mining, …
focus groups, usability panels, …
interviews, shoulder surfing, …
Prioritization
of identified need and derived requirements
User Requirements Evaluation and Prioritization (UREP) Working Group
• broad participation across architecture, deployment, operations, users, and service providers
Assessing
plans and deployments
through a variety of stakeholder-focused, facilitated workshops
• e.g., interactive ATAM sessions focused on identifying, quantifying, discussing tradeoffs
Representation
in the management of XSEDE
XSEDE Advisory Board
User Advisory Committee
Service Providers Forum