Visit additional Tabor Communication Publications
June 04, 2009
June 4 -- Less than three months after being given the go-ahead to upgrade their current facilities, the Swiss National Supercomputer Centre (CSCS) in Manno, Ticino, switched on the most powerful computer in Switzerland. The new supercomputer's peak performance of 141.6 teraflops is higher than all of the computer clusters in Switzerland put together.
The name of the supercomputer, "Monte Rosa," refers to the famous Swiss-Italian mountainous massif , and it is quite fitting that it carries connotations of conquering the peaks when you realise that barely three months elapsed between ETH Zurich giving the go-ahead to upgrade its predecessor "Piz Palu," and the new supercomputer being switched on. CSCS expects to declare the computer open for general access to its user community some time this week.
Quantum leap in supercomputing for Switzerland
In that short space of time, the Swiss National Computer Center made a quantum leap by its standards: the peak performance of the computer center increased more than eightfold. This makes CSCS one of the world's leading computer centers in high performance computing (HPC) in terms of computer capacity. At CSCS, they are confident that "Monte Rosa" will finish in the top fifty at the international supercomputer conference in Hamburg this June, where the "TOP500" fastest computers in the world will be announced. Should it do so, it would also be a major leap forward in high performance computing for Switzerland as a whole.
Today, computer simulation is widely considered to be the third pillar of science together with theory and experiment, and high performance computers are essential tools for many of the most advanced and insightful simulations. Nowadays it is commonplace to find that many practical experiments conducted in the lab are supported by numerical experiments carried out on a computer. These simulations offer scientists an insight into important intermediate stages in processes that would often be incomprehensible -- or even impossible to see at all -- under a microscope or using other experimental methods. Many of the simulations that are carried out on these large machines involve scenarios that cannot be carried out experimentally, such as studying the process of galaxy formation or determining the structure of minerals in the extreme temperatures and pressures at the centre of the Earth. Drafting the climate scenarios of the future would also be impossible without simulations carried out on supercomputers.
Preparing for reception
In the run-up to Monte Rosa's installation, the electrical infrastructure and the water-cooling system to support the new machine had to be created at CSCS in a very short time frame. Last week, the moment of truth finally came: an awning in front of CSCS sheltered the 30-ton cargo from the elements as it was delivered on four trucks. The twenty head-high cabinets that house the XT5 racks from the American company Cray were chaperoned by about a dozen electrical-engineers and computer construction specialists from the company, who worked round the clock to put the machine together.
Trials with "high-impact" projects
Almost a week on, 14,752 processors at CSCS are now capable of performing 141 trillion computer operations per second. This means that in one fell swoop CSCS has been able to increase its capacity enormously, and the scientific research community is now able to make use of this new machine, which is currently more powerful than all of the computer clusters in Switzerland put together. As a consequence, researchers have now been called upon to submit so-called high-impact projects involving challenging simulations that they have had in the pipeline, but which had simply not been feasible in Switzerland until now due to a lack of sufficient large scale computer facilities.
CSCS is looking to use these projects to test just how powerful the machine is. Thomas Schulthess, director of CSCS and professor of Physics at ETH Zurich, has already received the Gordon Bell Prize for similar test runs conducted last year with the supercomputer "Jaguar" in Oak Ridge (see ETH Life article from Nov. 21, 2008).
The next mountain already beckons
The CSCS team members are true summiteers always in search of the next peak to climb. No sooner had the new computer arrived than they were already planning the next upgrade. The national CSCS high performance computer center is aiming to provide its users with a petaflop computer capable of carrying out a quadrillion computer operations per second, giving it a peak performance similar to the machine that was inaugurated in Jülich, Germany, this week. "We are very enthusiastic about the possibility of pushing the boundaries of simulation-based science," explains Thomas Schulthess. He stresses just how important it was that HPC was declared a national strategy in Switzerland (see ETH Life interview from Dec. 17, 2008) and is happy that the Swiss University Conference (see ETH Life article from May 6, 2009), the state with its first stimulus package, and the ETH domain are backing the project and supporting it financially.
The Swiss National High-Performance Center (CSCS) was founded in 1991 by order of the state and has been run by ETH Zurich ever since. It provides its capacities for all national research institutions and thus works with both ETH institutes, the Swiss universities, the research establishments in the ETH domain, CERN and MeteoSwiss. The Monte Rosa supercomputer is funded by the state stimulus package, CSCS, grants from ETH Zurich and partly by the Paul Scherrer Institute (PSI), which has secured a share of the computer in the process.
Source: Swiss National Supercomputing Centre
In quieter times, sounding the bell of funding big science with big systems tends to resonate further than when ears are already burning with sour economic and national security news. For exascale's future, however, the time could be ripe to instill some sense of urgency....
In a recent solicitation, the NSF laid out needs for furthering its scientific and engineering infrastructure with new tools to go beyond top performance, Having already delivered systems like Stampede and Blue Waters, they're turning an eye to solving data-intensive challenges. We spoke with the agency's Irene Qualters and Barry Schneider about..
May 22, 2013 |
At some point in the not-too-distant future, building powerful, miniature computing systems will be considered a hobby for high schoolers, just as robotics or even Lego-building are today. That could be made possible through recent advancements made with the Raspberry Pi computers.
May 15, 2013 |
Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.