IBM Selected to Build World's Most Powerful Computing Grid
released
August 9, 2001
Contact
John Buscemi
jbuscemi@us.ibm.com
914.766.4495
Multiple IBM eServer* Intel-based Linux Clusters and Storage Connected by Superfast Network to Achieve 13.6 Trillion Calculations Per Second
ARMONK, NY IBM was today selected by a consortium
of four U.S. research centers to build the world's most powerful computing
Grid, an interconnected series of Linux(R) clusters capable of processing
13.6 trillion calculations per second. The Grid systemknown as the
Distributed Terascale Facility (DTF)will enable thousands of scientists
around the country to share computing resources over the world's fastest
research network in search of breakthroughs in life sciences, climate
modeling and other critical disciplines.
Funded by the National Science Foundation, the DTF is a joint undertaking
of the National Center for Supercomputing Applications (NCSA), the San
Diego Supercomputing Center (SDSC), Argonne National Laboratory, and the
California Institute of Technology. The Grid will include not only the
fastest supercomputers, but high-resolution visualization environments,
toolkits for Grid computing, and data storage facilities integrated into an
information infrastructure called the "TeraGrid."
Ushering in a new era of scientific collaboration, the Grid will allow
researchers to scan remote databases, run applications on far-flung
computers, and view complex computer simulations in real-time from widely
separated locations.
Unlike traditional supercomputers, which are typically housed at a single
location, Grids create vast pools of computing resources by connecting
multiple, often widely-distributed supercomputers using the Internet or
high-speed research networks as well as open source protocols from Globus
(http://www.globus.org/).
Organizations tap into these computing Grids to access
processing capacity, data storage and bandwidth in much the same way that
consumers draw electricity from a power grid.
IBM eServer, Storage and Services
IBM Global Services will deploy clusters of IBM eServer Linux systems at
the four DTF sites beginning in the third quarter of 2002. The serversdesigned
for the computational and data-intensive workloads run by both
commercial as well as technical customerswill contain the next
generation of Intel's Itaniium(TM) microprocessor, code-named McKinley.
IBM's world-renowned supercomputing software, CSM and GPFS, will handle
cluster and file management tasks. Myricom's Myrinet interconnect will
enable interprocessor communication.
The system will have a storage capacity of more than 600 terabytes of
data, or the equivalent of 146 million full-length novels. A substantial
portion of the Grid's storage infrastructure will be enabled by IBM
TotalStorage products and technologies.
The Linux clusters will be connected to each other via a superfast, 40
gigabit per second Qwest network, creating a single computing system able
to process 13.6 teraflops. A teraflop is a trillion calculations per
second. The DTF will be more than a thousand times faster than IBM's Deep
Blue supercomputer, which defeated chess champion Garry Kasparov in 1997.
Using open protocols, the Linux clusters will smoothly connect to a
heterogeneous collection of existing high performance computers at the four
labs, creating a giant virtual computer that may be accessed from any point
on the Grid.
"IBM's leadership in supercomputing technology and our commitment to Linux
and open standards enable us to provide the world's most powerful
computers," said David Turek, IBM vice president of Linux emerging
technologies. "Leveraging our strong relationship with Intel and Myricom in
conjunction with Qwest, we look forward to building a Grid computing system
that represents an important step in the evolution of the Internet and
scientific collaboration."
"The DTF will be the largest, most comprehensive infrastructure ever
deployed for scientific research, and it will be powered by IBM Linux
clusters," said Dan Reed, director of NCSA. "This is truly a cooperative
effort among research labs and major corporations such as IBM to develop a
new way to do computing and conduct research."
IBM Grid Expertise
IBM is the leading supplier of systems and services expertise to the
scientific and technical community. In addition to working with many of the
world's leading labs and research organizations in the development of Grid
projects, IBM Research used Globus technologies to build its own Grida
geographically distributed supercomputer linking IBM research and
development labs in the United States, Israel, Switzerland, Japan and
England. IBM's Global Services organization offers the complete range of
IT skills needed to build, run and maintain Grids.
To help customers manage complex Grids, IBM offers scalable supercomputing
systems and middleware with IBM eLiza self-management technologies. Project
eLiza, announced by IBM earlier this year, is a company-wide program to
develop systems that respond to the requirements of their environment in
order to optimize performance across a network, improve security and
survive failures.
IBM also plans to Grid-enable key IBM systems and technologies, allowing
them to be plugged into these growing worldwide networks quickly and
easily.
In the same way it played a leadership role in the commercial adoption of
Linux, IBM is working with the Globus open source development community and
the influential industry standards body, Global Grid Forum. Open protocols
are essential to Grids because they enable heterogeneous systems to work
together as a single system.
The National Center for Supercomputing Applications is the leading-edge
site for the National Computational Science Alliance. NCSA is a leader in
the development and deployment of cutting-edge high-performance computing,
networking, and information technologies. The National Science Foundation,
the state of Illinois, the University of Illinois, industrial partners, and
other federal agencies fund NCSA.
The National Computational Science Alliance is a partnership to prototype
an advanced computational infrastructure for the 21st century and includes
more than 50 academic, government and industry research partners from
across the United States. The Alliance is one of two partnerships funded by
the National Science Foundation's Partnerships for Advanced Computational
Infrastructure (PACI) program, and receives cost-sharing at partner
institutions.
SDSC is an organized research unit of the University of California, San
Diego, and the leading-edge site of the National Partnership for Advanced
Computational Infrastructure (NPACI). As a national laboratory for
computational science and engineering, SDSC is funded by the National
Science Foundation through NPACI and other federal agencies, the State and
University of California, and private organizations.
The National Partnership for Advanced Computational Infrastructure (NPACI)
is advancing science by creating a ubiquitous, continuous, and pervasive
national computational infrastructure: the Grid. This infrastructure for
the 21st century builds on dramatic advances in information technology to
enable distributed research by interdisciplinary teams. NPACI is funded by
the National Science Foundation and led by the San Diego Supercomputer
Center. It joins some 30 other funded partners and 16 domestic and
international affiliates and collaborates with the Alliance members on
numerous projects.
The nation's first national laboratory, Argonne supports basic and applied
scientific research across a wide spectrum of disciplines, ranging from
high-energy physics to climatology and biotechnology. Since 1990, Argonne
has worked with more than 600 companies and numerous federal agencies and
other organizations to help advance America's scientific leadership and
prepare the nation for the future. Argonne is operated by the University of
Chicago as part of the U.S. Department of Energy national laboratory
system.
Founded in 1891, Caltech has an enrollment of some 2,000 students, and a
faculty of about 290 professorial members. The Institute has more than
19,000 alumni. Caltech employs a staff of more than 2,400 on campus and
4,800 at JPL. Over the years, 28 Nobel Prizes and four Crafoord Prizes have
been awarded to faculty members and alumni. Forty-seven Caltech faculty
members and alumni have received the National Medal of Science; and eight
alumni (two of whom are also trustees), two additional trustees, and one
faculty member have won the National Medal of Technology. Since 1958, 13
faculty members have received the annual California Scientist of the Year
award. On the Caltech faculty there are 78 fellows of the American Academy
of Arts and Sciences; and on the faculty and Board of Trustees, 70 members
of the National Academy of Sciences and 46 members of the National Academy
of Engineering.
* The IBM eServer brand consists of the established IBM e-business logo
with the descriptive term "server" following it.
IBM, the e-business logo and xSeries are trademarks of IBM Corporation.
Intel is a registered trademark of Intel Corporation. Linux is a
registered trademark of Linus Torvalds. Other company, product and service
names may be trademarks or service marks of others.
Releases Archive