Keynote
Speakers
KEYNOTE
The Grid: Beyond the
Hype
Ian
Foster (Argonne National Laboratory, University of Chicago, USA)
Abstract
Grid technologies and infrastructure
support the integration of services and resources within and among enterprises,
and thus allow new approaches to problem solving and interaction within distributed,
multiorganizational collaborations. First developed and applied within eScience,
Grid tools are now increasingly being applied in commercial settings as well.
In this talk I discuss the current state and planned future directions for one
particular collection of Grid technologies, namely the open source Globus Toolkit.
I examine, in turn, the underlying requirements that motivate its design; the
components that it provides for security, resource access, resource discovery,
data management, and other purposes; the protocol and interface standards that
underlie its design and implementation; the ecosystem of complementary tools,
deployments, and applications that build on and/or complement these components;
and the nature of the most urgent challenges that must be overcome to expand
its utility and breadth of application.
Dr. Ian Foster is Senior Scientist and Associate Director of
the Mathematics and Computer Science Division at Argonne National Laboratory,
Professor of Computer Science at the University of Chicago, and Senior Fellow
in the Argonne/U.Chicago Computation Institute. He has published five books
and over 200 papers and technical reports in parallel and distributed processing,
software engineering, and computational science. He currently co-leads the Globus
project with Dr. Carl Kesselman of USC/ISI, which was awarded the 1997 Global
Information Infrastructure ``Next Generation'' award and which provides protocols
and services used by many distributed computing projects worldwide. He also
co-leads the GriPhyN and Earth System Grid projects which are extending and
applying Grid concepts in challenging application domains, and the GRIDS Center,
which is developing a national middleware infrastructure. He co-founded the
influential Global Grid Forum and co-edited the book “The Grid: Blueprint for
a New Computing Infrastructure.”(the first edition, the second edition) DOE,
DARPA, NSF, NASA, and Microsoft support his research. Foster is a fellow of
the American Association for the Advancement of Science and of the British Computer
Society, which recognized his work on the Strand parallel programming language
with its 1989 award for technical innovation.
KEYNOTE
High Performance Computing
Trends, Supercomputers, Clusters and Grids
Jack
Dongarra (University of Tennessee, Oak Ridge National Laboratory, USA)
Abstract
In this talk we will look at how
High Performance computing has changed over the last 10-year and look toward
the future in terms of trends. In addition, we advocate the `Computational Grids'
to support `large-scale' applications. These must provide transparent access
to the complex mix of resources - computational, networking, and storage - that
can be provided through aggregation of resources.
Jack Dongarra earned a Bachelor of Science in Mathematics from Chicago
State University in 1972. A year later he finished a Master of Science in Computer
Science from the Illinois Institute of Technology. By this time, he was already
involved in the EISPACK project producing high quality, portable, Fortran implementations
of state-of-the-art algorithms for numerical linear algebra. He formally received
his Ph.D. in Applied Mathematics from the University of New Mexico in 1980.
He worked at the Argonne National Laboratory until 1989, becoming a senior scientist.
He now holds an appointment as University Distinguished Professor of Computer
Science in the Computer Science Department at the University of Tennessee, is
an Adjunct R&D Participant in the Computer Science and Mathematics Division
at Oak Ridge National Laboratory (ORNL), and an Adjunct Professor in Computer
Science at Rice University.
He specializes in numerical algorithms
in linear algebra, parallel computing, use of advanced-computer architectures,
programming methodology, and tools for parallel computers. His research includes
the development, testing and documentation of high quality mathematical software.
He has contributed to the design and implementation of the following open source
software packages and systems: EISPACK, LINPACK, the BLAS, LAPACK, ScaLAPACK,
Netlib, PVM, MPI, NetSolve, Top500, ATLAS, and PAPI. He has published approximately
200 articles, papers, reports and technical memoranda and he is co-author of
several books. He is a Fellow of the AAAS, ACM, and the IEEE and a member of
the National Academy of Engineering.
KEYNOTE
Making Grids Real:
Standards and Sociology
Charlie
Catlett (Argonne National Laboratory, USA)
Abstract
Grid technologies
attempt to answer several fundamental needs. Complex applications often require
resources (instruments, computers, information...) that exist in multiple locations
under the administration of multiple organizations. Grid technology attempts
to address the combining of these resources to fulfill the needs of these advanced
applications. However, there is a second fundamental need that Grid technology
aims to meet-providing for these organizations to work together without loss
of autonomy or security. Because effective Grid systems require software and
information technology from multiple sources, standards are essential. Because
Grid systems require resources from multiple organizations, careful process
and sociology structures are necessary. These include process, policy, and organizational
incentives. Catlett will talk about lessons learned as Chair of Global Grid
Forum and as Executive Director of the US National Science Foundation's TeraGrid
project in these areas.
Charlie Catlett is a Senior Fellow at the Argonne National Laboratory,
Executive Director of the TeraGrid project, Chair of Global Grid Forum, and
the Director of the I-WIRE optical network consortium.
Charlie was network architect in
the original TeraGrid proposal and became Executive Director when funding was
awarded in late 2000. TeraGrid is a 90M dollar NSF-funded project that is deploying
a 25 Teraflops computational Grid system integrating resources at Argonne, Caltech,
NCSA, PSC, and SDSC.
Since 1999 Charlie has chaired the
Global Grid Forum. The several thousand GGF participants come from over 30 countries
and some 400 organizations. As GGF's first chair, Charlie led the development
of GGF's processes, organization, governance, and culture and established a
not-for-profit company to support GGF's activities. Charlie also serves as editor
of the GGF document series.
In 1999 Charlie became Director
of the I-WIRE project, a 12M dollar dark fiber and WDM transport network initiative
that has deployed an optical network among 10 sites in the Chicago area. I-WIRE
provides dark fiber and lambda services to projects including Starlight, the
40 Gb/s NSF TeraGrid Backplane network, and the NSF-funded Optiputer project.
Prior to joining Argonne in 2000,
Charlie was Chief Technology Officer at the National Center for Supercomputing
Applications (NCSA). With Larry Smarr, Charlie co-authored a seminal paper in
1992, "Metacomputing," in the Communications of the ACM, which contributed
to the concept of Grid computing. That same year Charlie's paper "In Search
of Gigabit Applications," published in IEEE Network, received the Fred
W. Ellersick award for best paper in an IEEE journal. His most recent publication
is "Standards for Grid Computing: Global Grid Forum," in the inaugural
issue of the Journal of Grid Computing.
KEYNOTE
e-Science and Web
Services Grids
Tony
Hey (University of Southampton, UK)
Abstract
The talk will introduce the concept
of e-Science and briefly describe some of the main features of the ￡250M 5 year UK e-Science Programme.
This review will include examples of e-Science applications not only for science
and engineering but also for e-Health and the e-Enterprise. The importance of
data curation will be emphasized and the move towards 'Web Services Grids' -
Grid middleware based on open standard Web Services.
Tony Hey is Professor of Computation at the University of Southampton
and has been Head of the Department of Electronics and Computer Science and
Dean of Engineering and Applied Science at Southampton. From March 31st 2001,
he has been seconded to the EPSRC and DTI as Director of the UK's Core e-Science
Programme. He is a Fellow of the British Computer Society, the Institution of
Electrical Engineers and the Institution of Electrical and Electronic Engineers.
Professor Hey is European editor of the journal 'Concurrency and Computation:
Practice and Experience' and is on the organising committee of many international
conferences.
Professor Hey has worked in the
field of parallel and distributed computing since the early 1980's. He was instrumental
in the development of the MPI message-passing standard and in the Genesis Distributed
Memory Parallel Benchmark suite. In 1991, he founded the Southampton Parallel
Applications Centre in 1991 that has played a leading technology transfer role
in Europe and the UK in collaborative industrial projects. His personal research
interests are concerned with performance engineering for Grid applications but
he also retains an interest in experimental explorations of quantum computing
and quantum information theory. As the Director of the UK e-Science Programme,
Tony Hey is currently excited by the vision of the increasingly global scientific
collaborations being enabled by the development of the next generation 'Grid'
middleware. The successful development of the Grid will have profound implications
for industry and he is much involved with industry in the move towards OpenSource/OpenStandard
Grid software.
Tony Hey is also the author of two
popular science books: 'The Quantum Universe' and 'Einstein's Mirror'. Most
recently he edited the 'Feynman Lectures on Computation' for publication, and
a companion volume entitled 'Feynman and Computation'.
KEYNOTE
About Grid
Greg
Astfalk (Hewlett-Packard)
Abstract
This talk's title is purposely simplistic.
We want to use the opportunity to discuss Grid in a very realistic, broad-reaching,
and practical way. Grid's most important role is largely misunderstood, as is
its ETA (Estimated Time of Arrival).
We'll consider the key enablers
for Grids and the mega-trends that are acting as forcing functions on Grid.
A key point is that Grid, in the future, is largely about SOAs (Service Oriented
Architectures) and less so about scientific computing.
We'll discuss its fit with commercial
IT, with robust management of IT resources, including services themselves, and
of the need for open standards.
Above all, it is hoped this talk
will be provocative, illuminating, and thought-provoking.
Bio
Greg Astfalk is Chief Scientist
in the Office of Corporate Strategy and Technology. His responsibilities are
in helping to set HP's technology directions and agenda. He has held several
positions with HP. Among them was Chief Scientist for the Technical Computing
Division. Astfalk has been with Hewlett-Packard since its acquisition of Convex
Computer Corporation in 1995.
At Convex Astfalk was the Staff
Scientist. His responsibilities were in developing and analyzing parallel algorithms
and as a consultant to the MPP hardware designers.
Before Convex he was at Bell Laboratories
for 17 years. While there he did computational mathematics in subject areas
including plastics extrusion, lasers, mathematical optimization, thermal stresses,
IC encapsulation, crystal growth, and others.
Astfalk has been an editor for SIAM
(Society for Industrial and Applied Mathematics) for 12 years. He edits a monthly
column entitled "Applications on Advanced Architecture Computers."
KEYNOTE
Many Faces of Grid
Computing
Greg
Rankich (Microsoft)
Abstract
High Performance Computing and Grid
Computing are among the fastest growing server workloads worldwide, driven by
enterprises and academic adoption. HPC features some of the most demanding and
exciting application scenarios that drive innovation in distributed system development,
large scale management, parallel computing, networking and storage. This talk
will take a hard look at the various forms of HPC and Grid Computing and attempt
to distill hype from reality. The talk will compare and contrast high performance
computing clusters, desktop scavenging, "data-grids", and cooperative
grids and how each approach can be used.
Microsoft's view and efforts in this space will also be discussed.
Greg Rankich - As senior product manager in the Windows Server Division,
Greg helps drive aspects of marketing, product management and business planning
for Microsoft's High Performance Computing efforts. Greg is also responsible
for our strategic partnerships with corporate and academic organizations, including
Microsoft's investments in Cornell Theory Center. Prior to taking on the HPC
role, Greg led the initial effort, from conceptual plans to managing the worldwide
launch of Windows 2000 Datacenter Server and the Windows Datacenter Program.
During his 5-year plus career at Microsoft, Greg has led joint efforts with
Compaq/HP, IBM, and Dell to further Windows penetration into the Enterprise
market. Greg holds a B.A. in International Business and an M.B.A. from Washington
State University.
KEYNOTE
From Science to Enterprise
- Intel's Grid activities
Karl
Solchenbach (Intel)
Abstract
The presentation addresses the current
status of UNICORE and will describe Intel$B!G(Bs concepts for a new Grid Programming
Environment (GPE). UNICORE is now heavily used in academia and industry; the
presentation will show some examples of successful application enabling with
UNICORE. This includes results of the EUROGRID project as well as work of other
Grid projects in and outside Europe. The UNICORE architecture is based on three
layers: the client interfaces to the end-user, providing functions for creating
jobs to be run and monitoring their progress, the gateway is the single entry
point for all users at a centre and cooperates with the firewalls to only allow
legitimate users in, and the server schedules UNICORE jobs for execution, translating
the specification to the commands appropriate for the target platform. The server
also controls transmission of files, including output and result files that
can be later retrieved by the user.
Currently UNICORE is modified to
support the new Grid standards OGSA and WS-RF. In addition, Intel extends the
successful graphical UNICORE client to a next generation Grid Programming Environment
(GPE). When the focus of the Grid applications will change from science to enterprise
applications, Grid-specific environments and development tools will be needed.
GPE will make it very easy to use the Grid, it will support all Grid standards
and Grid middleware systems like Globus, UNICORE and others. With the GPE development
Intel will accelerate the grid-readiness of scientific, technical and enterprise
applications in various vertical markets.
Karl Solchenbach
received a diploma in mathematics from the University of Bonn in 1980.
From 1977 to 1985, Karl Solchenbach
worked in Federal Institute for Mathematics and IT (GMD). He Specialized on
numerical solvers for partial differential equations, multigrid algorithms and
computational fluid dynamics applications, parallel algorithms for distributed–memory
multiprocessor systems. Since 1986 he is head of application software development
in SUPRENUM. After 1991, he is co-founder and managing director of Pallas. Pallas
has developed to a leading HPC software company with a strong focus on Grid
technology. When leading Pallas, he is active in Grid and is co-initiator of
the Unicore project and coordinator of the EUROGRID project.
Since 2003, he joined Intel and
is Engineering Manager of the Intel Cologne Software Lab, which leads the development
of cluster tools and grid technology in Intel and is involved in various Grid
projects (NextGrid, SimDat, UniGrids).
KEYNOTE
Dawning4000A: a 11Tflops
Linux Cluster
Ninghui
Sun (Institute of Computing Technology, CAS, China)
Abstract
Dawning4000A is a 11Tflops Linux
Cluster, which is developed by the National Research Center for Intelligent
Computing Systems (NCIC) and Dawning Corporation. It is a grid-enabling high
performance computer, and is used as the capability power station of the China
National Grid (CNGrid). Dawning4000A
is powered by 2560 AMD Opteron microprocessors, and with supportive of grid-enabling
components, such as GridKVM, GridKey, GridRouter, GridView, GridGateway.
Dr. Ninghui Sun, is the
director of National Research Center for Intelligent Computing Systems (NCIC)
at Institute of Computing Technology (ICT) of Chinese Academy of Science (CAS).
He graduated at Peking University in 1989, and got the Master and PhD degrees
from ICT of CAS in 1992 and 1999
respectively. Dr. Sun is the architect and main designer of Dawing1000, Dawning2000,
Dawning3000, Dawning4000 high performance computers.
KEYNOTE
On demand Grid Infrastructure
Initiatives
Moon
(James) Kim (IBM, USA)
Abstract
IBM's on demand computing represents
a fundamental shift in how clients use technology. On demand systems should
be powered by responsive, flexible, focused, and resilient technologies. Grid
technology is a key component of on demand computing. It uses open standards
so different systems can work together and link with devices and applications
across organizational and geographic boundaries. And, it's virtualized to hide
the physical resources from the application level, making the best use of technology
resources and minimizing complexity for users.
An on demand computing approach
enables a network of organized, self-managing computing components to deliver
what the client needs, when they need it, with less effort. The dynamics of
the market are changing and this requires a new way of thinking about business
processes and supporting IT infrastructures.
IBM is entering the next phase of
e-business on demand, where companies need to move beyond simply integrating
their various processes, to a world in which they can respond to fluctuating
market conditions in real time. IBM is leading the way to help our clients become
on demand company and the grid technology will play a critical part of this
effort. In this presentation, some of Grid initiatives are introduced.
Dr. Kim is a Senior Technical Staff Member and a Master Inventor of
IBM Corp. He, as a Chief Architect for Strategic Infrastructure Development
in IGA, is responsible for developing on demand grid infrastructures and for
follow-on next generation computing system. He has developed many system solutions
such as the Advanced Enterprise Information Search & Delivery System for
ibm.com Internet and Broadband Interactive System. He was also a key architect
and developer of IBM S/390 families, MPP, HPPS systems.
He received 11 patents and filed more than 15 inventions.
He has authored numerous books, R & D papers, inventions publications that
are related to the systems design.
KEYNOTE
Dell’s HPCC Strategy
and solutions
Eric
D'Angelo (Dell)
Abstract
This discussion will over topics
on Dell’s strategy to commoditize clustering technologies and bring the Dell
model to High Performance Compute Clusters.
Topics include:
n
History of HPCC
n
Trends in the market (DELL'S FOCUS)
n
HPCC Architecture (DELL'S HPCC SOLUTIONS)
n
Components of a cluster (INTERCONNECT, SERVER,STORAGE,etc)
n
Dell’s Blade strategy (DELL 1855MC )
n
The top500 list (included dell's positons)
n
Dell’s Cluster research team (ET team, solution center)
Eric D'Angelo is a Senior
Server Marketing Manager for Dell Asia Pacific, focusing on Software and Solutions.
Eric has been with Dell for over 4 years in numerous roles including Enterprise
Technologist for Infrastructure. Eric has more than 17 years experience in information
technology field. Eric also has a background in Civil Engineering.
KEYNOTE
Enabling Grid collaboration
in Academic research with technology partnership
Andrew
Lim (Sun Microsystems)
Abstract
Grid computing is about collaboration
and partnership between commerical application and academic research is a critical
driving force for technology evolvement. In Asia Pacific, Sun's partnership
with academic institutions has evolved into a APSTN community
(Asia Pacific Science and Technology Netowrk ) with focus on scientific
research, and knowledge sharings.
Andrew has more than 17-years
of IT experience. He is currently managing Sun's education and research in Asia
South, as well as leading a team in engaging universities and research institutions
in APAC in the areas of Grid and HPTC...Andrew has been in Sun for more than
8yrs, and had held various leadership position in Singapore and the region.