• July 2018
    M T W T F S S
    « Jun    
  • Advertisements

European Supercomputer Shines With Energy Efficiency eeNews Europe

Christoph Hammerschmidt
June 27, 2018

As Europe seeks to compete globally in developing innovative supercomputer architectures, the Julich Supercomputing Center (JSC) in Germany has launched JUWELS (Julich Wizard for European Leadership Science), marking significant progress in a new generation of flexible, modular supercomputers. JUWELS ranked 23rd on the most recent TOP500 supercomputer list, making it the fastest German system. JSC’s Thomas Lippert believes the modular design is critical to an affordable and energy-efficient technology that can aid forthcoming exascale systems. Lippert’s adaptable design concept, known as “Smart Exascale,” provides for a supercomputer with several specialized modules that can be dynamically and flexibly combined via software. The cluster module’s Intel Xeon 24-core Skylake CPUs allow for a theoretical peak performance of up to 12 petaflops, which equals the approximate computing power of 60,000 PCs. The module has extremely energy-efficient hot water cooling, allowing it to cool the majority of the waste heat with hot water directly with the outside air without additional cooling generators. Next year, JUWELS will receive a booster module for massively parallel applications that will multiply computing power. The cluster module reached a computing speed of 6.2 petaflops in initial test runs based on the Linpack benchmark.



Move Over, China: U.S. Is Again Home to World’s Speediest Supercomputer

The New York Times
Steve Lohr, June 8, 2018

Oak Ridge National Laboratory’s Summit system is expected to be named the world’s fastest supercomputer in the Top500 rankings coming out later this month, effectively ending China’s five-year reign at the top of the list, says the University of Tennessee’s Jack Dongarra. Summit can execute calculations of 200 petaflops a second, and is expected to help expedite the development of artificial intelligence and other technologies. Summit is more than twice as fast as the leading supercomputer ranked in the most recent Top 500 list in November, which is based at China’s National Supercomputing Center. Powering the Oak Ridge system are 9,216 central processors from IBM and 27,648 graphics processors from Nvidia interconnected by 185 miles of fiber-optic cable. Summit is viewed as a stopgap measure, as supercomputers five times faster are in the works both in the U.S. and overseas.


UT Austin’s New Supercomputer Stampede2 Storms Out of the Corral in Support of U.S. Scientists

UT News
Faith Singer-Villalobos
July 28, 2017

The University of Texas at Austin’s (UT Austin) Texas Advanced Computing Center (TACC) has launched Stampede2, the most powerful supercomputer at any U.S. university, which UT Austin president Gregory L. Fenves says will enable researchers “to take on the greatest challenges facing society.” Stampede2 was built with a $30-million National Science Foundation (NSF) grant, and its applications will include large-scale models and data analyses using thousands of processors simultaneously, and smaller computations or interactions via Web-based community platforms. TACC executive director Dan Stanzione predicts Stampede2 “will serve as the workhorse for our nation’s scientists and engineers, allowing them to improve our competitiveness and ensure that UT Austin remains a leader in computational research for the national open science community.” Stampede2 will have a peak performance of 18 petaflops while consuming half the power of Stampede1. It will be made available to researchers via NSF’s Extreme Science and Engineering Discovery


Google Plans to Demonstrate the Supremacy of Quantum Computing

IEEE Spectrum, Rachel Courtland
May 24, 2017

Google researchers say they plan to boost the volume of superconducting qubits built on integrated circuits (ICs) to create a 7×7 array and push operations to the limits of even the best supercomputers, demonstrating “quantum supremacy” by year’s end. The team says it will perform operations on a 49-qubit system that will trigger chaotic evolution yielding what appears to be random output, which classical computers can model for smaller systems. University of California, Santa Barbara professor John Martinis says the qubits constituting the array also could be employed to build larger “universal” quantum systems with error correction, capable of performing useful tasks such as decryption. Martinis says the challenge of scaling up the quantum IC involves maintaining qubits’ function without losing fidelity or boosting error rates. “Error rate and scaling tend to kind of compete against each other,” he notes. The team also sees the possibility of scaling up systems beyond 50 qubits without error correction.



A Scientist and a Supercomputer Re-create a Tornado

UW-Madison News, Eric Verbeten, March 13, 2017

Researchers at the University of Wisconsin-Madison are using supercomputer simulations to study the structure of tornado-producing supercell thunderstorms. The researchers say they can create in-depth visualizations of supercells and discern how they form and ultimately spawn tornadoes. The most recent simulation recreates the “El Reno” tornado, which touched down in Oklahoma in 2011 and caused damage over a 63-mile area. Using real-world observational data, the researchers were able to recreate the weather present at the time of the storm and witness the steps leading up to the creation of the tornado. The simulation reveals in high resolution the numerous “mini-tornadoes” that form at the onset of the main tornado. As the funnel cloud develops, the mini-tornadoes begin to merge, adding strength to the storm. The researchers used the Blue Waters Supercomputer housed at the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign.


The Next Supercomputing Superpower–Chinese Technology Comes of Age

Asian Scientist (01/03/17) Rebecca Tan

China has been the ranking leader on the Top500 list of the world’s most powerful supercomputers since June 2013, claiming unsurpassed growth compared to all other countries, according to University of Tennessee professor Jack Dongarra. The ascent of China, which did not even make the Top500 list until 2001, raised fears of its supercomputers being used for nuclear applications, given the growing need for such resources to simulate nuclear tests. Despite a U.S. ban on selling microchips to China, Stony Brook University professor Deng Yuefan says China’s supercomputing progress has continued unabated. One result was the rollout of China’s Shenwei SW26010 chips, which put the Sunway TaihuLight system at the top of the Top500 list with a Linpack benchmark of 93 petaflops and also tripled its predecessor’s efficiency. Deng says China is making investments in software development to put its supercomputers to good use. He says this is evident in the use of Sunway TaihuLight by three of the six finalists for the 2016 ACM Gordon Bell Prize, including the winning team, at the SC16 conference in November. Meanwhile, China also is in a race with Japan and the U.S. to build the first exascale supercomputers.


U.S. Exascale Computing Update With Paul Messina

HPC Wire (12/08/16) Tiffany Trader

In an interview, Distinguished Argonne Fellow Paul Messina discusses stewardship of the Exascale Computing Project (ECP), which has received $122 million in funding (with $39.8 million to be committed to 22 application development projects, $34 million to 35 software development proposals, and $48 million to four co-design centers). Messina notes experiments now can be validated in multiple dimensions. “With exascale, we expect to be able to do things in much greater scale and with more fidelity,” he says. Among the challenges Messina expects exascale computing to help address are precision medicine, additive manufacturing with complex materials, climate science, and carbon capture modeling. “The mission [of ECP] is to create an exascale ecosystem so that towards the end of the project there will be companies that will be able to bid exascale systems in response to [request for proposals] by the facilities, not the project, but the typical DOE facilities at Livermore, Argonne, Berkeley, Oak Ridge, and Los Alamos,” Messina says. In addition to a software stack to meet exascale app needs, Messina says there should be a high-performance computing (HPC) software stack “to help industry and the medium-sized HPC users more easily get into HPC.” Messina also stresses the need for exascale computing to be a sustainable ecosystem.