The WUN Grid

1
The WUN Grid
An infrastructure for collaborative research in
the Worldwide Universities Network
Presented by David De Roure and Allison Clark
2
Grid Computing

• Roots in high performance computing and
  specialized scientific problem-solving
• Grid computing has emerged as a powerful general
  purpose infrastructure to enable new research and
  learning
• Its contemporary definition by Foster Kesselman
  is
• A Grid brings together core grid computing
  infrastructure services, applications and users

Coordinated resource sharing and problem solving
in dynamic, multi-institutional virtual
organizations
3
International Grid Scene

• The Global Grid Forum brings theinternational
  community together
• UK and European activities emphasize the Semantic
  Grid, which promotes all aspects of
  interoperability
• UK e-Science program attracts attention forbeing
  applications-led and multidisciplinary
• Investment for sustainable infrastructure
  evident
• e.g. NSF Middleware Initiative, OMII
• In practice there are many grids
• organizational barriers impede creation of
  general purpose international Grids

4
WUN and Grid Synergy

• WUN will benefit from new collaborative research
  enabled by applications on the Grid
• The WUN Grid will benefit from the organizational
  infrastructure provided by WUN
• WUN easily overcomes institutional barriers which
  constrain other Grids
• WUN Grid is competitive against other Grid
  exercises
• Hence WUN Grid offer significant enhancement for
  WUN with prospect of high impact, competitive
  research
• Gives WUN an identifiable infrastructure and a
  unique platform for basis of funding applications

5
Excellent circumstances

• WUN partners include international leaders in
  Grid computing
• e.g. San Diego Supercomputer Centre at UCSD,
  National Centre for Supercomputing Applications
  at UIUC, key Grid software from Wisconsin,
  CiteSeer at Penn State
• All UK WUN partners are major players in
  e-Science, with significant international
  leadership
• White Rose Grid provides track record in creating
  a multi-institutional grid
• Many WUN Grand Challenges will benefit from Grid
  computing
• WUN Grid is itself a Grand Challenge and it
  supports other WUN Grand Challenges

6
Strategy

• Move towards vision of a WUN Grid which
• Creates new, high-impact research
• Generates IP and learning enhancements
• Generates revenue
• Is sustainable
• Consult users and identify priority areas where
  WUN Grid is positioned to make an impact
• Create an implementation plan for WUN Grid,
  balancing data, collaboration and computation
• Implement a foundation WUN Grid and example
  grassroots applications to inform the WUN Grid
  roadmap

7
Informatics Group Progress

• WUN researchers met in San Francisco December
  2002, hosted by Sun
• Subsequent discussions at Global Grid Forum
• Application priorities
• Arts and humanities
• Social Sciences
• Infrastructure priorities
• Data grid
• Collaborative Grid
• Computational Grid

• Second meeting in Santa Clara December 2003,
  hosted by Sun
• Agreed governance structure
• Created infrastructure and applications teams
• Have produced
• What document
• How document
• Prototype grid
• Work has gone to schedule WUN works

8
Foundation of the WUN Grid

• SDSC
• Manchester
• Southampton
• White Rose
• NCSA
• A functioning, general purpose international Grid

Manchester-SDSC mirror
9
WUN Grid Portal
10
Arts and Humanities

• Emerging activity in US and UK
• Link with HASTAC - Humanities, Arts, Science, and
  Technology Advanced Collaboratory
• strategic alliance of scientists, humanists,
  artists, social theorists, legal specialists, and
  information technology specialists
• Link with UK Arts and Humanities Research Board
  and Arts and Humanities Data Service
• Link with GGF
• Humanities, Arts and Social Sciences Research
  Group
• Workshop on Social Factors, Humanities, Arts and
  Social Sciences Old Challenges and New
  Disciplines for Grid Computing

11
HASTAC Humanities, Arts, Science and Technology
Advanced Collaboratory

• How much the scientist can learn from the
  humanist and artist and vice versa.
• Humanist can add the why to the gee whiz part
  of the technology. Historians and philosophers on
  one side of campus. Computer scientist and
  engineers on the other. No more build it and they
  will come.
• The challenge is the building of bridges among
  diverse cultures and communities technology,
  humanist, artists, social scientist.
• Speak a common language

12
HASTAC Founding Members

• University of California Humanities Research
  Institute
• Maryland Institute for Technology and the
  Humanities (MITH)
• Virginia Institute for Advanced Technology in the
  Humanities
• Duke University's John Hope Franklin Center and
  Humanities Institute
• Center for Information Tech Research in the
  Interest of Society (CITRIS)

• California Digital Library
• Stanford Humanities Lab
• Florida International University
• CAL IT²
• University of Illinois at Urbana Champaign
• National Center for Supercomputing Applications
  at the University of Illinois (NCSA)
• San Diego Supercomputer Center at the University
  of San Diego (SDSC)

13
HASTAC maybe Mission Statement

• HASTAC is an international, interdisciplinary
  consortium which seeks to create, develop,
  advance and utilize a broad range of leading
  computing and information systems while
  contributing to an understanding of the
  interconnections between the human sciences,
  natural sciences, arts, and technology in a
  complex global society. HASTAC, in partnership
  with the science and technology communities, is
  dedicated to the creation and development of
  humane technologies and technological humanism.

14
Cyberinfrastructure Enables People
Scientists, Engineers, Decision Makers, Policy
Makers, Media and Citizens Engaging in
discovery, analysis, discussion, deliberation,
decisions, policy formulation and communication
Collaboration Framework facilitates Idea and
Knowledge Sharing, eLearning and Multi-Objective
Decision Support Processes
Analysis Framework facilitates Data and Model
Discovery, Exploration, and Analysis via the
Collaboration Framework
Data Management Framework builds logical maps of
distributed, heterogeneous information resources
(data, models, tools, etc.) and facilitates
their use via the Analysis and Collaboration
Frameworks
Physical Infrastructure
Courtesy of Tom Prudhomme, NCSA
15
Music demonstrator

• Activities underway
• Access Grid being enhanced for musical
  performance
• Collaboration between music information retrieval
  experts at UIUC and WUN Grid
• Semantic Web collaboration tools, including
  capture and replay, being applied to musical
  performance
• Example
• Convert digitized musical recording into musical
  score
• Requires polyphonic pitch transcription
  (computational challenge)
• Use this for search in music information
  retrieval (data challenge)
• Use for re-synthesis

16
Accessing musical content
17
Polyphonic pitch transcription
18
WUN collaborative Grid

• Proven technology from
• UK Advanced Knowledge Technologies
  Interdisciplinary Research Collaboration (won
  Semantic Web Grand Challenge)
• Link with CiteSeer (Giles) at Penn State
• Proposing AKTive Seer to run on WUN Grid as
  collaboration tool for WUN
• Supports communities of practice
• Matchmaking between WUN experts
• Infrastructure challenge
• Involves integrating terabytes of data

19
AKTive Seer
20
Social Sciences

• Working with georeferenced data
• links to DialogPlus (NSF/JISC)
• Urbana, UCSD, Manchester, hold massive data
  collections
• Taking advantage of the e-social science
  opportunities
• Demonstrator (Leeds) Hydra International
• Finds groups of similar cities across UK, US,
  France and Norway
• e.g. could be used in a comparative analysis of
  planning policies between similar cities across
  international boundaries

21
Hydra
22
Geospatial digital library

• From Santa Clara meeting
• Builds on DialogPlus
• Discussing using Southampton tools with
  Alexandria Digital Library content

23
Earth Science Grand Challenge

• Long-term potential for using WUN Grid
• Fibre-optic links ashore followed by high-speed
  high-bandwidth communications provide live video
  feed from hydrothermal vent-sites to the
  classroom
• Command capability provides pan and tilt controls
  on video cameras, continuous monitoring of
  temperature changes etc.
• Remote manipulation using robot arm

24
Summary

• Creation of WUN Grid offers very significant
  benefits to WUN in terms of synergising resources
  for research impact
• Considerable progress in 3 months
• Roadmap agreed by WUN Informatics Group in Santa
  Clara
• Arts and humanities application area have held
  meetings in US and UK and are moving forward
• Foundation WUN Grid infrastructure in place
• Social sciences making rapid progress
• Collaborative tools moving forward

25
Contact

• David De Roure, University of Southampton
• dder_at_ecs.soton.ac.uk
• http//www.ecs.soton.ac.uk/dder
• See also www.wungrid.org

26
Interoperability
SemanticWeb
SemanticGrid
Scale of Interoperability
ClassicalWeb
ClassicalGrid
Scale of data and computation