GLORIAD: USA-Russia Lightpath Enables Fast Data Transfer of Terabyte-sized Scientific Datasets
Knoxville/Oak Ridge, Tennessee, USA (June 1, 2007)
The week of May 6, scientists from the National Center for Data Mining (NCDM) at the University of Illinois at Chicago and the Geophysical Center at the Space Research Institute, Russian Academy of Sciences, Moscow, demonstrated a new method for distributing extremely large volumes of scientific information across the world. They successfully moved 1.4 TeraBytes (TB) of data in about 4.5 hours over a 1 Gbps lightpath between Chicago and Moscow as part of the Teraflow Network initiative. This event, which represents the highest performance information transfer ever recorded between these two countries, was made possible by a unique international organizational partnership.
Although the amount of science information is growing rapidly, the ability to move it on the regular Internet is still very limited. NCDM partnered with Russia’s Geophysical Center to demonstrate a new capability for moving science data by transferring the Sloan Digital Sky Survey (SDSS) dataset between their two sites, using a specialized international communications facility.
Using NCDM’s open-source, high-performance network transport protocol UDT (UDP-based Data Transfer) on the Teraflow Network, researchers were able to quickly transfer the SDSS astronomy catalog data, between Chicago and Moscow. The 2.5 TB catalog is compressed to 1.4 TB, split into 60 files and, is distributed to astronomers around the world from the NCDM in Chicago. Using UDT, the 1.4 TB was transferred over a 1 Gbps lightpath and then decompressed in Moscow to its original size. It now resides on a local server <www.skyserver.ru> in Moscow.
This data transfer had a sustained rate of 711 Mbps and a peak rate of 844 Mbps, and took about 4.5 hours to complete. This is about the speed that the data could be moved across the city of Chicago over a 1Gbps network, which graphically illustrates how barriers of distance are being eliminated by the new communications infrastructures and technologies. These techniques are required for research and experimentation for many science disciplines, and in the future it may also be used for many types of data intensive commercial applications.
This accomplishment was made possible through a unique partnership among organizations in eleven countries that have created advanced communication facilities at locations literally around the world. GLORIAD, the Global Ring Network for Advanced Applications Development, is a consortium of several countries, notably the USA, Russia, China, Korea, Canada, the Netherlands, and the Nordic countries (Denmark, Sweden, Norway, Finland and Iceland), that are contributing networking capabilities to build a global 10 Gbps optical network around the northern hemisphere of the globe in supportof advanced science and engineering. In the USA, GLORIAD is supported by the National Science Foundation’s International Research Network Connections (IRNC) program, which also funds a 10 Gbps path between Chicago and Amsterdam called TransLight/StarLight. GLORIAD has been provided with a 3 Gbps path on TransLight/StarLight to allow a direct high-performance connection between the USA and Europe.
GLORIAD’s Russian partners recently installed a 10 Gbps path from Amsterdam to Moscow, provided by the Russian Research Center “Kurchatov Institute.” This allowed a 1 Gbps lightpath to be dedicated to the Teraflow Network, from Chicago (the StarLight facility), to Amsterdam (the NetherLight facility) and then on to Moscow (the MoscowLight facility). GLORIAD participants are part of a global initiative called the Global Lambda Integrated Facility (GLIF), which promotes the paradigm of lightpaths, or lambda networks, for data-intensive scientific research and applications.
This science demonstration was also supported by NCDM’s Teraflow Network, an international facility designed to develop innovative technologies to stream massive distributed datasets over high-performance networks, at 1 Gbps, 10 Gbps and multiple 10 Gbps. The TeraFlow Network is being used as a next-generation platform, capable of supporting data-intensive applications, including many requiring information transfers that cannot be supported by traditional networks. The TeraFlow Network is developing techniques that will be required by future global applications.
“This is the latest in a string of demonstrations that proves that it is now practical for the working scientist to efficiently access terabyte size datasets from anywhere in the world. All it takes are today’s high-performance networks and new network protocols, such as UDT,” said Robert Grossman, NCDM director at the University of Illinois at Chicago. “With the technology now available, there is no reason for scientists not to have access to the latest data available in order to advance their research.”
“We look forward to using these new technologies to share and mine very large databases in global change, space weather and remote sensing studies,” said Mikhail Zhizhin, head of the Telematics Lab at the Geophysical Center in Moscow, “and to applying the technologies from the Teraflow Network to the larger GLORIAD infrastructure. In particular, the Research Group is working with the USA National Geophysical Data Center (NOAA) on the Space Physics Interactive Data Resource (SPIDR), and is working with Microsoft Research Cambridge on the Environmental Scenario Search Engine (ESSE). Additionally, there is strong demand to transmit real-time data streams and high-resolution images, which has not previously been possible. ”
”This is a significant achievement between USA and Russian scientists,” stated Alexey Soldatov, co-director of GLORIAD/Russia and director of the Institute of Information Systems, Russian Research Center “Kurchatov Institute” (RRC “KI”). GLORIAD/Russia, based at RRC “KI”, provides support and development of the networking infrastructure for scientist and educators. “In addition, our Research Center is one of the leaders of the nationwide Russian Data Intensive Grid program, that will use GLORIAD's advanced networking infrastructure to support data-intensive projects and frontier experiments in high-energy physics, nanotechnology, gravitational wave research, digital astronomy and molecular genomics."
“The ability to move multi-terabyte datasets internationally in a matter of hours, and ultimately minutes, has been based on the cooperation and efforts among many international teams and it builds a solid foundation for future international science projects,” said Natalia Bulashova, GLORIAD/USA co-principal investigator.
“Lessons learned on the Teraflow Network can be expanded to the entire GLORIAD community, and ultimately other GLIF international partners, said Greg Cole, GLORIAD/USA principal investigator. “No matter how fast we increase capacity and services on the GLORIAD network, the various science groups out there are moving faster. It’s a real challenge for us, but it’s a good challenge.”
The GLORIAD/USA team worked with teams from Russia (GLORIAD member - the Russian Research Center “Kurchatov Institute”; MoscowLight/RBNET/Russian Institute for Public Network (RIPN); Institute of Space Research at the Russian Academy of Sciences; the Geophysical Center at the Russian Academy of Sciences; and, the Visualization Laboratory at Moscow State University); the USA (National Center for Data Mining at the University of Illinois at Chicago; the International Center for Advanced Internet Research at Northwestern University; TransLight/StarLight; StarLight; Johns Hopkins University; and the GLORIAD UT-ORNL JICS); and, the Netherlands team (GLORIAD member - NetherLight/SURFnet; SARA; and the University of Amsterdam).
# # # # #
The GLORIAD (Global Ring Network for Advanced Application Development) advanced science internet network was launched in January 2004 by the USA, Russia and China, expanded its reach in 2005 – to Korea, Canada and The Netherlands – and in 2006 to the five Nordic countries of Denmark, Sweden, Finland, Norway and Iceland. GLORIAD provides an optical network ring encircling the northern hemisphere of the globe with individual network circuits providing up to 10 Gbps – promoting new opportunities for cooperation for scientists, educators and students. The GLORIAD project is supported by the Ministry of Science and Education of Russian Federation, the National Science Foundation of USA, the USA Research & Education (R&E) network National LambdaRail, the Chinese Academy of Science, the Ministry of Science and Technology of Korea, Canadian non-profit association CANARIE, the national R&E network of Netherlands SURFnet, the national R&E network of the Nordic Countries NORDUnet, as well as a number of other organizations representing countries which participate in the project. GLORIAD/USA is based at the University of Tennessee – Oak Ridge National Laboratory, Joint Institute for Computation Science. GLORIAD/Russia is based at the Russian Research Center “Kurchatov Institute”. GLORIAD/Netherlands is based in Amsterdam and managed by SURFnet. For more information, see <www.gloriad.org>.
About the Teraflow Network
The Teraflow Network (TFN), under the leadership of the National Center for Data Mining (NCDM) at the University of Illinois at Chicago and the International Center for Advanced Internet Research (iCAIR) at Northwestern University, is both: 1) a unique infrastructure enabling scientists to share large scientific datasets with their collaborators around the world; and, 2) an international experimental facility designed to create and test advanced networking technologies, new protocols and services, and integrated middleware to support distributed, high-performance data-intensive applications. The TFN supports 1 Gbps, 10 Gbps and multiple 10 Gbps data streams and connects computer clusters distributed over three continents. The TFN allows scientists to contribute terabyte datasets to repositories as shared resources – enabling colleagues around the world to use them as the basis for new discoveries. With the TFN, it is easy to explore, integrate and analyze datasets contributed by colleagues. The TFN is sponsored in part by the National Science Foundation of the USA, the U.S. Army, the U. S. Naval Research Laboratory, NASA, and the University of Illinois at Chicago. For more information, see <www.teraflowtestbed.net>.
About The Geophysical Center, Russian Academy of Sciences
The Geophysical Center of the Russian Academy of Sciences was founded in 1958 by a special Decree of the USSR Council of Ministers with the mandate to archive and exchange data resulting since the first International Geophysical Year. Its primary goal is to conduct basic and applied research in the fields of geophysics and geoinformatics. For more information, see <www.gcras.ru/index_e.html>
About The RRC “Kurchatov Institute”
The Russian Research Center “Kurchatov Institute”, Russia’s leading research and development institution in the field of nuclear energy, was founded in 1943. The Kurchatov Institute is funded through the Ministry of Education and Science of the Russian Federation and through international cooperation and commercial projects. It is divided into 15 institutes and six scientific and technological divisions. The Institute conducts research on controlled thermonuclear fusion, plasma physics, solid state physics, superconductivity, molecular physics, physical and inorganic chemistry, chemical physics, chemistry, safety of new technologies, ecology and health, biology, genomics and bioinformatics, biotechnology, element basis of microelectronics, material science, nanotechnology, networking and information science. The RRC “KI” is the founder of Russia’s Internet and the Russian Research and Education network, and the Russian Institute for Public Networks (RIPN). GLORIAD/Russia is based at the RRC “KI” and the Russian Institute for Public Networks (RIPN). For more information, see <http://www.kiae.ru>
About The Russian Institute for Public Networks (RIPN)
Russian Institute for Public Networks (RIPN) was founded in 1992 by the Science and Higher School Committee of Russia, Russian Research Center "Kurchatov Institute" and the Computer Center of the Kurchatov Institute. RIPN operates the Russian Backbone Network and MoscowLight facility, supports Technical Center of national domain TLD .RU and develops public Internet Exchanges in the largest cities of Russia. RIPN is GLORIAD/Russia’s network operator. For more information, see <http://www.ripn.net>
Russian Backbone Network (RBNet) (since 1996) is a backbone that serves regional, specialized and corporate networks for Research and Education. At present, it is a high-speed IP network with Points-of-Presence (POPs) in all Federal Regions of Russia, and is migrating to a Hybrid Network Infrastructure.
2006-January: Open the GLIF Open Lambda Exchange (GOLE) Facility, MoscowLight, to support the Hybrid Network Infrastructure
2007-January: Establish 10 Gbps link from Moscow (MoscowLight) to Amsterdam (NetherLight)
2007-April: Establish lightpath between Moscow, Amsterdam and Chicago
2007-2008: Plan to establish 2.5Gbps - Moscow - Khabarovsk link, with a connection to Hong Kong (HKOEP-HKLight) and Korea (KRLight).
For more information, see < http://www.rbnet.ru >
About The Institute of Space Research, Russian Academy of Science
Institute of Space Research, Russian Academy of Sciences (RAS), is the leading RAS organization in scientific research areas such as: Outer Space, Solar System planets and other objects of the Universe. The Space Research Institute (IKI) is primarily in charge of long-range planning and elaboration of space research programs, of which a considerable part is performed within the framework of international space research cooperation. For more information, see <http://www.iki.rssi.ru/eng/index.htm>.
About The National Center for Data Mining at the University of Illinois at Chicago
The National Center for Data Mining (NCDM) at the University of Illinois at Chicago (UIC) was founded in 1998 as a national resource for high-performance and distributed data mining. The Center performs research, coordinates standards, operates testbeds, and engages in outreach. The Center coordinates the development of the Predictive Model Markup Language (PMML), a standard for statistical and data mining models, and operates the Teraflow Network, a network for distributing large e-science datasets. For more information about NCDM, see <http://www.ncdm.uic.edu>.
In cooperation with USA and European National Research & Education Networks, TransLight/StarLight is implementing a strategy to best serve established USA and European production science, including usage by those scientists, engineers and educators who have persistent large-flow, real-time, and/or other advanced application requirements. TransLight/StarLight currently provides two connections between the USA and Europe for production science: a routed connection that connects the pan-European GÉANT2 to the USA Internet2, National LambdaRail and ESnet networks, and a switched connection that is part of the LambdaGrid fabric being created by participants of the Global Lambda Integrated Facility (GLIF) and used by the GLORIAD community and the Teraflow Network project. Major funding is provided by the USA National Science Foundation International Research Network Connections (IRNC) program, award OCI-0441094 to the University of Illinois at Chicago (UIC), for the period February 2005 - January 2010. For more information, see <http://www.startap.net/translight>.
About StarLight (sm)
StarLight is an advanced optical infrastructure and proving ground for network services optimized for high-performance applications. StarLight is the GLIF Open Lightpath Exchange (GOLE) in Chicago. Operational since summer 2001, StarLight has 1GE and 10GE switch/router facilities and true optical switching for wavelengths. StarLight is being developed by the Electronic Visualization Laboratory (EVL) at the University of Illinois at Chicago (UIC), the International Center for Advanced Internet Research (iCAIR) at Northwestern University, and the Mathematics and Computer Science Division at Argonne National Laboratory, in partnership with Canada's CANARIE and the Netherlands' SURFnet. See < www.startap.net/starlight>. StarLight (sm) is a service mark of the Board of Trustees of the University of Illinois.
About UT-ORNL JICS
The University of Tennessee (UT) Oak Ridge National Laboratory (ORNL) established the Joint Institute for Computational Sciences (JICS) to advance scientific discovery and state-of-the-art engineering and to further knowledge of computational modeling and simulation by taking full advantage of the terascale and beyond computers in the National Center for Computational Sciences housed at ORNL, and by educating a new generation of scientists and engineers well-versed in the application of computational modeling and simulation for solving for the most challenging scientific and engineering problems.
GLORIAD/USA based at UT-ORNL JICS. For more information, see <http://www.gloriad.org/gloriad/team/usa/index.html> and <http://www.jics.utk.edu/jics_about.html>
NetherLight is the GLIF Open Lightpath Exchange (GOLE) in Amsterdam, The Netherlands and has been operational since January 2002. It is an advanced open optical infrastructure and proving ground for network services optimized for high-performance applications, especially those that require Grid infrastructure. NetherLight provides multiple specialized international high-performance communication services in partnership with other advanced research networks. NetherLight has supported multiple landmark data-intensive science experiments and demonstrations and is pioneering new concepts for architecture that may find their way into other GOLEs worldwide and the traditional telecommunications world as they move to next-generation architectures. SURFnet enables breakthrough education and research. SURFnet develop and operate the national SURFnet6 network and provides innovative services in the areas of security, authentication and authorization, group communication and video. SURFnet is a partner in SURF, the collaborative organization in which research universities, universities of applied sciences and research institutions aimed at innovative ICT facilities operate at a national and an international level. GLORIAD/Netherlands is based in Amsterdam and managed by SURFnet. For more information, see <http://www.netherlight.net/info/about/introduction.jsp>, <http://www.surfnet.nl> and <http://www.gloriad.org/gloriad/team/netherlands/index.html >
Joe Mambretti, Robert Grossman, Yunhong Gu, Thomas DeFanti, Maxine Brown, Alan Verlo, Linda Winkler, Alex Szalay, Greg Cole , Natalia Bulashova, Evgeny P.Velikhov, Alexey Soldatov, Alexey Platonov , Mikhail Zhizhin, Ravil Nazirov, Anton Korotin, Alexander Ilin , Michael Boyarsky, Veniamin Konoplev, Olga Starostina, Kees Neggers,, Cees de Laat,, Erik-Jan Bos, Paola Grosso, Bram Peeters, J.P. Velders, Wounter Huisman, John Lankford, Peter de Boer, Thomas Tam, Bill St.Arnaud , Rene Hatem , NetherLight NOC , RBnet NOC, StarLight NOC.
For more information:
Co-Principal Investigator, GLORIAD