October 11, 2011
URBANA-CHAMPAIGN, Ill., Oct. 10 -- The Extreme Science and Engineering Discovery Environment (XSEDE), the National Science Foundation cyberinfrastructure project that replaces and expands on the NSF TeraGrid, has transitioned its network backbone infrastructure to use the FrameNet service of National LambdaRail (NLR). This change, a switch from older technology used by TeraGrid, provides a more flexible infrastructure and saves substantial costs for connecting XSEDE service providers.
"We've eliminated the need for the TeraGrid-owned capital equipment," said Linda Winkler, who leads the XSEDE networking group, "making a more direct interconnect between XSEDE sites. This will save hundreds of thousands of dollars in equipment cost, as well as tens of thousands of dollars in annual maintenance."
XSEDE has transitioned to the NLR FrameNet service, which is part of the nationwide optical-fiber infrastructure of NLR, a consortium of more than 280 U.S. universities and private and government laboratories that provides high-performance network services for research and education. XSEDE replaced SONET/OC192 circuits with 10-GigE (10 gigabits per second Ethernet) connectivity to FrameNet for all core XSEDE service provider sites. XSEDE has a dedicated 10-GigE circuit from NLR between backbone nodes in Chicago and Denver. As with the predecessor TeraGrid service, all XSEDE participants have full site-to-site connectivity —including broadcast capability — between all the service provider sites.
The XSEDE networking group, led by Winkler (of Argonne National Laboratory in Chicago), spent the past few months switching the backbone infrastructure of the TeraGrid to the newer, more cost-effective technology. Each of the core XSEDE sites did the switch at a separate time to minimize impact to users. "As far as I know," said Winkler, "nobody noticed that we did the change, and in my estimation that's a job well done."
The changed infrastructure provides flexibility to easily extend the XSEDE network as additional sites are included. "FrameNet has the advantages of a private network along with a rich set of connectivity options," added Winkler. "The footprint of FrameNet nodes across the country makes it feasible to extend the XSEDE network backbone to other cities and to add capacity to existing sites."
"National LambdaRail is pleased to provide high-performance networking connectivity for the XSEDE partners and the National Science Foundation," said Peter O'Neil, NLR vice president of research. "NLR was founded with the goal of supporting data-intensive computational science and research, and now, with long-term financial support in place, we are re-committed to this mission."
In coming months, XSEDE networking staff will implement performance monitoring services that will make it easier for researchers to obtain optimum network performance. In addition, networking staff will research alternative connection strategies, including the ability to take advantage of dynamic circuits on demand.
The eight core XSEDE service-provider sites directly affected by the network transition are: Indiana University, National Center for Supercomputing Applications, National Institute for Computational Sciences, National Center for Atmospheric Research, Pittsburgh Supercomputing Center, Purdue University, San Diego Supercomputer Center, and Texas Advanced Computing Center.
More about XSEDE: https://xsede.org.
More about NLR: http://www.nlr.net.
XSEDE, the most advanced, powerful, and robust collection of integrated digital resources and services in the world, is a single virtual system that scientists can use to interactively share computing resources, data, and expertise. The five-year, $121 million project is supported by the National Science Foundation, and it replaces and expands on the NSF TeraGrid project.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?