SC12 News: Internet2 Delivers 320 Gigabits, NVIDIA Launches Tesla K20X Accelerator
November 13th, 2012 By: John Rath
Internet2 community members, including NASA, NOAA and leading universities, will demonstrate this week at the Supercomputing Conference how the advanced Internet2 Network supports high-energy physics research, such as the search for the “Higgs boson” particle and climate and atmospheric science for improved predictions of extreme weather, like Hurricane Sandy. To support the scientists at SC12, Internet2 deployed three 100 gigabit Ethernet links to its new Software Defined Networking (SDN)-based Advanced Layer 2 Service, as well as two 10 GE links to its Research and Education IP backbone for a record-shattering capacity of 320 gigabits from a single provider. With a recently deployed upgrade the Internet2 network will be the first open, national-scale production network to use SDN and OpenFlow standards.
“Breakthrough discoveries in atomic physics, climate science, and cancer research require the most advanced technologies available,” said Internet2 President and CEO Dave Lambert. “Internet2’s unwavering commitment to providing the capabilities to our community that meet the rapidly evolving demands of research has never been stronger or more capable. We look forward to providing ongoing support to NASA, NOAA and their university-based research counterparts as our community continues to accelerate new global discoveries, together.”
The Large Hadron Collider (LHC) produces data flows of more than 100 petabytes per year and relies on the efficient movement of data sets between global sites. At SC12 industry partners will help demonstrate an inter-connect with three major LHC Tier-2 computing sites and the SC12 show floor using 100 Gbps technology (and all of Internet2’s Advanced Layer 2 Service links to SC12) through a collaboration between Caltech, the University of Victoria and the University of Michigan.
NVIDIA Launches GPU Accelerators
NVIDIA (NVDA) unveiled the NVIDIA Tesla K20 family of GPU accelerators and the technology powering Titan, the world’s fastest supercomputer according to the November 2012 Top500 list. The flagship product of the accelerated computing product line, the Tesla K20x accelerator, is based on the NVIDIA Kepler computer architecture. It provides 3.95 teraflops single-precision and 1.31 teraflops double-precision peak floating point performance.The Tesla K20 accelerator offers 3.52 teraflops of single-precision and 1.17 teraflops of double-precision peak performance.
“We are taking advantage of NVIDIA GPU architectures to significantly accelerate simulations in such diverse areas as climate and meteorology, seismology, astrophysics, fluid mechanics, materials science and molecular biophysics,” said Dr. Thomas Schulthess, professor of computational physics at ETH Zurich and director of the Swiss National Supercomputing Center. “The K20 family of accelerators represents a leap forward in computing compared to NVIDIA’s prior Fermi architecture, enhancing productivity and enabling us potentially to achieve new insights that previously were impossible.”
Using Tesla K20X accelerators, Oak Ridge’s Titan achieved 2,142.77 megaflops of performance per watt, which surpasses the energy efficiency of the number 1 system on the most recent Green500 list of the world’s most energy-efficient supercomputers.
NVIDIA also announced that 70 more widely used applications have added support for GPU acceleration so far this year, bringing the total number available to researchers, engineers and designers to more than 200. The newest applications to offer GPU acceleration include ANSYS Fluent, MSC Nastran, and CHARMM.