|
|
 |
KEYNOTE SPEAKERS
|
|
Title:
From E-Science to Grid Computing and Beyond
Speaker: Thomas Fahringer
Institute of Computer Science
University of Innsbruck, Austria
Abstract
The Grid community has made substantial progress
in the area of Grid computing, and yet most success
stories are largely limited to e-science. In part, expectations
have been raised high by claims that the Grid can meet
today's needs for generic, planet-wide distributed computing
and delivery of results. Many of these claims have been
demonstrated through pilot projects which unfortunately
have not achieved lasting influence -- in contrast to
what has been envisioned by some Grid researchers.
This presentation will go back to the
roots of Grid computing: E-science. We will concentrate
largely on the objective to create an invisible Grid
for application users and developers.
Many existing Grid application systems
provide the application developer with a non-transparent
Grid. Commonly, application developers are explicitly
involved in selecting software components deployed on
specific sites, mapping applications onto the Grid,
or selecting appropriate computers for their applications.
Moreover, many programming environments are either implementation-
and technology-specific, or they force the developer
to program at a low-level middleware interface.
For a case study, this talk will describe
the ASKALON Grid application development and computing
environment, whose ultimate goal is to provide an invisible
Grid to application users and developers. Furthermore,
this talk will make the case of an interesting alternative
research path with the objective to automatically generate
workflows based on semantic technologies. Finally an
example centered around the development of online-games
will be used to illustrate how methods originating from
scientific and Grid computing can be applied to meet
the needs of industry applications. This example offers
renewed hope that future
for the Grid may still be bright.
About the
Speaker
Dr. Thomas Fahringer is a Professor of Computer
Science at the University of Innsbruck in Austria. He
is leading a research group in the area of distributed
and parallel processing which develops the ASKALON system
to support researchers worldwide in various fields of
science and engineering to develop, analyse, optimize
and run parallel and distributed scientific applications.
Before joining the University of Innsbruck,
Dr. Fahringer worked as an Associate professor at the
University of Vienna where his research focused on compiler
technology and tools for high performance applications.
Dr. Fahringer is a graduate of the Technical University
of Vienna with a doctorate in computer science.
Dr. Fahringer was involved in numerous
national and international research projects including
PPPE, Apart, EGEE 1-2, ASG, K-Wf Grid, CoreGrid and
edutain@grid which were all funded by the European Union.
His group currently coordinates two EU projects (edutain@grid,
EC-GIN). Fahringer has published 3 books, 30 journals
and magazines and more than 100 reviewed conference
papers including 3 best/distinguished IEEE/ACM
papers.
|
|
Title: The New e-Science
Speaker: Professor David De Roure
School of Electronics and Computer Science,
The University of Southampton, UK.
Abstract
In their early years, e-Science and
cyberinfrastructure were dominated by heroic science
using heroic infrastructures. But now we are seeing
researchers across all disciplines taking advantage
of new technologies to do new research. Much of this
user-centred activity is drawing on the Web as a distributed
application platform, with mashups for integration,
easy access to resources "in the cloud", and
social networking to share the pieces and practice of
digital science. As in other walks of life, the new
technologies are empowering the individual, a trend
set to continue with the increasing power of the multicore
desktop. This evolution throws some of our existing
approaches, epitomised by the Grid, into question. Adopting
a user and application perspective, this talk will present
a definition of the New e-Science. It will also discuss
how we might create a flourishing ecosystem of scientists,
software developers and service providers rather than
just a pipeline of
provision.
About
the Speaker
Prof David De Roure leads the e-Science
activities in the School of Electronics and Computer
Science, University of Southampton. Closely involved
with the UK e-Science programme since its inception,
he has worked with many disciplines from science and
social science to arts and humanities. He is Chair of
the Open Middleware Infrastructure Institute UK. An
advocate for Web in e-Science, he leads the Semantic
Grid Research Group in the Open Grid Forum where he
is also an e-Science Area Director, and has led two
workshops this year on Web 2.0 and the Grid. He is a
member of the Scientific Council of the Web Science
Research Initiative.
|
|
| Title:
Reputation Studies – A Grassroots Approach and its
Implications for e-Science and Grid Computing
Speaker: Professor Xiaoming Li
Director of Institute of Network Computing and Information
Systems, School of Electronic Engineering and Computer
Science Peking University, Beijing, China
Abstract
Reputation is an important issue, and has been
of interest to many researchers from different areas,
including e-Science and Grid computing. This talk will
address a specific problem, namely how different reputation
models may be mapped onto real systems. This will be
exemplified by our work based on Maze, a large-scale,
non-commercial, peer-to-peer file sharing system deployed
in China by the network lab at Peking University. The
talk will first address some aspects of user behaviors
in peer-to-peer systems such as Maze. Using the logs
collected from Maze, three typical kinds of “negative”
behaviors are identified and investigated, namely free
riding, whitewashing, and collusion. In addition, we
have tested the effectiveness of EigenTrust (a famous
P2P reputation algorithm) against real data and have
found that EigenTrust has some difficulties in generating
proper trust values for certain peers. I will present
some measures to resolve these problems.
About The Speaker
Professor LI Xiaoming received his Ph.D. in
Computer Science from Stevens Institute of Technology
(USA) in 1986 and has since taught at Harbin Institute
of Technology and Peking University. He is the director
of Institute of Network Computing and Information Systems
at Peking University. Under his direction, the institute
founded and has been operating the Chinese web archive
Webinfomall (http://www.infomall.cn), the search engine
Tianwang (http://e.pku.edu.cn), the peer-to-peer file
sharing network Maze (http://maze.pku.edu.cn), the distributed
lecture video-on-demand system Realcourse (http://realcourse.grids.cn),
the programming online judge system Bailian (http://poj.grids.cn),
and other popular web channels. He is a member of Eta
Kappa Nu, a senior member of IEEE, currently a Vice
President of Chinese Computer Federation, International
Editor of Concurrency (USA), and Associate Editor of
Journal of Web Engineering (Australia). He has published
over 100 papers, authored Search Engine – Principle,
Technology, and Systems (Science Press, 2005), and received
numerous achievement awards from the Ministry of Science
and Technology, Beijing Municipal Government, Ministry
of Education, and other agencies.
|
|
Title: Building the Indian
Grid Computing Agenda
Speaker S. Ramakrishnan
Director General
Centre for Development of Advanced Computing
Abstract
Governments and user-community in Europe, USA and
elsewhere in the world have worked on large and small
projects for about a decade now in Grid Computing
for scientific community, which has now come to be
known as e-science. As the next generation cyber-infrastructure
for scientific community, it is viewed with great
hope and hype! Enterprise vendors and start-ups are
also waiting, after some early announcements. Standards
process has also been on. And parallel efforts in
Web have also progressed, converging with Grid standards
efforts. Clearly the eco-system has been emerging,
but the critical mass is yet to develop. Local context
of each country or community or region has also influenced
building of large scale initiatives.
Indian success stories in many sectors in the last
decade are too well known – IT, Telecom including
mobile, manufacturing as in the case of Auto sector,
Finance sector including banking & stock markets,
skilled work force, global investments -- to name
a few. Clearly economy is booming and there is emphasis
on India’s future as a knowledge economy.
From building India’s first supercomputer in
1991, and moving on to successive generations of PARAM
systems and facilities since then, C-DAC has been
working with the High Performance Computing user community
in different fields ranging from Atmospheric sciences
and Life Sciences to Earth Sciences, Material Sciences,
Astrophysics etc. While various users in academia
& R&D have been adding HPC systems, more recently
at least three systems from this community have appeared
in top 500 list, with one from CRL ranked # 4. Mission
agencies and Industries are also beginning to realize
the importance of benefitting from HPC systems.
Since end of 2004, C-DAC has embarked on a national
grid computing project, funded by Department of IT,
Govt. of India, Garuda, with 40 other premier academic
and research institutions as partners through a Proof-of-Concept
phase. India has also been an active player in the
CERN initiative with BARC, TIFR and many other institutions
playing eminent role over the last four years. Together,
EU-India project has also enabled collaboration between
communities across India & Europe.
Given the above, there are many creative conditions
to mount a concerted national initiative to build
the Indian Grid Computing agenda.
However, any such exercise in presenting a business
case has to necessarily contend with:
· How does one fit it in the midst of the
cacophony of viewpoints in the evolving grid computing
technology context?
· Why should anybody sit up and listen? What
can we promise as outcomes and in what time-frames?
· How do we build constituencies or low hanging
fruits?
The talk will address many of the above issues keeping
the unique context of India in mind, even while benefitting
from global experiences and global collaboration.
About The Speaker
Srinivasan Ramakrishnan is currently
Director General of Centre for Development of Advanced
Computing (C-DAC).
Ramakrishnan was Founder Project
Director of ERNET, the National Academic and Research
Network, a collaborative effort of five IITs, IISc,
former NCST and DIT (former DoE). ERNET brought Internetworking
into the country during 1986-96 and pioneered many
milestones in the field of computer networking. Over
800 institutions and 100,000 users were connected
by 1996 before the initiative led to a society, ERNETIndia
in 1998. It was
rated as one of the most successful UNDP projects
in the world. More recently, he represented Government
of India at the "Global forum on Internet Governance"
held during 25-27 March, 2004 at New York under the
auspices of UN ICT Task Force. He did his B. Tech
and M. Tech in Electronics from Indian Institute of
Technology, Madras wherein he
was a merit scholar and Siemens Prize winner. Later
he did PGDBAfrom MDI. He has traveled widely and has
presented a number of papers in national and international
fora.
|
|
Title:
IBM initiatives in the area of Deep Computing and Green
IT
Speaker: Dr.P .Sambath
Narayanan
Abstract:
Computing and Green IT are the
two major trends in the computing industry today. IBM
has been successfully driving research in these areas.
In this talk we will be covering the details of these
activities including technology details and breakthroughs.
Specifically this talk will cover
Power processors, Blue Gene and power virtualization,
Record breaking performance and scalability of some
of the benchmarks and applications would be presented.
This talk will also cover IBM initiatives
in the area of Green IT. More specifically what is Green
IT, What is IBM doing in terms of research and product
development and how Humanity in general and customers
in particular are benefiting from the Green IT program.
About The Speaker
PhD from IIT Delhi
Currently working with IBM STG lab services, IBM India
as HPC and Data Centre Consultant. Prior to IBM, he
has Worked in organizations like CDAC, SGI, & Sun
Microsystems.
In his role as a technical evangelist he has been covering
the AP region
He was instrumental in setting up Technology Centres
for Sun.
Has been in the HPC industry for the past 16 years.
Has been actively involved in Parallel application development
on variety of massively parallel super computers and
Shared memory supercomputers.
His another area of specialization include Data management.
As an IT specialist , he has architected enterprise
data management solutions for major banks and Telcos
in India.
He is a regular speaker in technology forums in the
areas of Grid, HPC and Data management.
|
|
EU-IndiaGrid
Presentation
Speaker: Shri P Dhekne
Abstract
The EU-IndiaGrid project, funded by the European Commission
under the Research Infrastructures Programme, aims to
make available a common, interoperable Grid infrastructure
to the European and Indian Scientific Community, in
order to support existing EU-Indian collaborations in
eScience and promoting new ones.
The major objectives of EU-IndiaGrid can be summarized
as follows:
- To offer an effective answer to the demanding computing
needs of several common EU-India research projects,
supporting the interconnection and interoperability
of EGEE with the Indian Grid infrastructure for the
benefit of eScience applications;
- To identify and aggregate actual and potential user
communities for the newly implemented common infrastructure
and promote its use through a specific outreach and
dissemination programme;
- To develop synergies between scientific research and
industrial communities, in order to foster the deployment
of grid techniques in research and industrial applications
within the Indian subcontinent.
As of September 2007, after just one year of activity,
the project has already taken significant steps towards
supporting the interconnection and interoperability
of the major European and Indian Grid infrastructures.
Key achievements include:
- Setting up a test-bed for the benefit of EU-India
Grid applications and mobilising a hardware infrastructure
of about 1200 core processors and 50 TB of disk for
the benefit of EU-India Grid applications.
- Creating the conditions for the access to a high speed
(1 Gb/s) EU-India intercontinental link for the EU-IndiaGrid
research Community thanks to the support of Government
of India Department of Atomic Energy.
- Providing Indian researchers with access to the intercontinental
Grid Infrastructure, using a temporary Certification
Authority through cooperation with Academia Sinica Taiwan
(EGEE partner and Regional Operation Centre for Asia)
- Taking concrete steps towards an internationally recognised
Indian Certification Authority with C-DAC, spearhead
of the GARUDA Indian National Grid Initiative, acting
as the selected institution.
- Building a Networked Community of over 450 members
from more than 300 individual organisations with very
highly qualified players from academic and research
institutions, government agencies and commercial organisations.
Transferring knowledge of Grid tools to researchers
and scientists, allowing them to use more advanced computing
tools and to port applications onto the Grid. Three
EU-IndiaGrid Workshops were successfully held in Bangalore,
Mumbai and Pune, combining Training Tutorials and Networking
Sessions. An additional training event dedicated to
gLite site managers was organised in Kolkata. The Conference
in Bangalore in December 2007, organised in cooperation
with the 3rd IEEE Conference on e-science and grid computing,
marks another key networking end knowledge-sharing event
for the project. |
|
|
