A-Z Index | Directory | Careers

ESnet Collaborates with Dutch SURFnet, Scandinavian NORDUnet to Create Network Research Framework

March 9, 2010

Energy Sciences Network (ESnet), the high speed network connecting thousands of Department of Energy scientists and managed by Lawrence Berkeley National Laboratory, is embarking on a cooperative research effort with SURFnet (www.surfnet.nl), the national computer network for higher education and research in the Netherlands, and Scandinavian research network NORDUnet (www.nordu.net) to explore potential synergies between the unique technical knowledge and expertise that each organization has to offer.

As the pace of global collaborations increases, scientists need to reliably exchange massive datasets between research centers. ESnet is working with SURFnet and NORDUnet to explore common methodologies for reserving end-to-end bandwidth to accelerate such data transfers. This capability is essential to provide the high speed connectivity that can support "exascale science," which requires data on the petascale and intensive computing to conduct visualization, simulation, modeling and analyses.

ESnet has a history of successful international and industry collaborations, the PerfSONAR collaboration with Internet2 and U.S. and European national research and educational networks, and Fenius with the Global Lambda Integrated Facility www.glif.is

Keeping up with the pace of technical innovation required to advance today's science is a difficult task for any single institution. "By pooling our expertise with SURFnet and NORDUnet for open source development of OSCARS and DRAC and making it available for participation from the wider community, we’ll be able to develop services more quickly and efficiently," said Steve Cotter, head of ESnet. "This particular effort was developed to address the need to build dynamic multilayer virtual circuits in order to deliver integrated IT services necessary for researchers to exchange the vast volumes of scientific data involved in global scientific collaborations."

 “This collaboration will add a new layer of functionality and ease of use to the existing research networking infrastructure," said Kees Neggers, managing director of SURFnet. "The open source nature of the software development and the use of GLIF Open Lightpath Exchange Points to interconnect our networks will make it easy and transparent for any others to join the collaboration later on.”

"The collaboration with ESnet and SURFnet will integrate these high level network provisioning technologies with tools such as DRAC and OSCARs to address the needs of exascale applications such as large scale data distribution and management and globally distributed work flow models." said Rene Buch, CEO of NORDUnet.

Shared research strengthens the networking community
The collaborative efforts will initially focus on achieving seamless multidomain interworking using different implementations of bandwidth reservation and provisioning software. ESnet's goal is to establish a governance framework to support a vibrant open-source community which can develop software that interoperates across multiple domains. An important project goal is to provide the software framework to support network research and the next generation of network virtualization. While ESnet, NORDUnet, and SURFnet are sharing best practices from their research into the standardization process of the Open Grid Forum, another objective is to integrate these standards back into the appropriate software to validate them and enable wider adoption in global networks.

Merging scheduling, design and virtualization agents
OSCARS, ESnet's automated scheduling and provisioning software, enables researchers to get the right-size network capacity to transfer vast datasets generated by large scientific instruments in fields including plasma physics, computational astronomy, combustion research and climate science. Users with an ESnet account can reserve capacity on the network and OSCARs guarantees the bandwidth necessary to transfer data within the specified time window. OSCARS was developed with funding and support from DOE's Office of Advanced Scientific Computing Research.

Nortel's DRAC software acts as a virtualization agent, configuring an optimal end-to-end network out of diverse sections. DRAC intelligently allows applications to control their demand for network resources without forcing them to interact directly with evolving network protocols. By provisioning necessary resources and steering data through the network either dynamically or on a reservation basis, DRAC enables users to optimize network capabilities, while reducing operations and investment costs. DRAC was originally developed as a commercial service mediation middleware portable to any JAVA platform, so it is a robust platform for open source development.

NORDUnet's MANTICORE enables network engineers to lay out design constraints for core network engineering and automated tools to do detail allocation and configuration of routers in order to streamline configuration of core network addressing, inter- and intra-domain routing protocols, and peering policy. MANTICORE was developed as a collaborative effort within the European community with funding from the European Commission, NORDUnet, and corporate sponsors.


About Computing Sciences at Berkeley Lab

High performance computing plays a critical role in scientific discovery. Researchers increasingly rely on advances in computer science, mathematics, computational science, data science, and large-scale computing and networking to increase our understanding of ourselves, our planet, and our universe. Berkeley Lab’s Computing Sciences Area researches, develops, and deploys new foundations, tools, and technologies to meet these needs and to advance research across a broad range of scientific disciplines.