Hurricane Gustav. New advanced computing abilities will allow scientists to better understand the links between weather and climate. (Credit: NOAA)



Petascale Climate Modeling Heats Up


The development of powerful supercomputers capable of analyzing decades of data in the blink of an eye mark a technological milestone capable of bringing comprehensive changes to science, medicine, engineering, and business worldwide.

Researchers at the University of Miami's Rosenstiel School of Marine and Atmospheric Science, collaborating with NCAR (National Center for Atmospheric Research), COLA (Center for Ocean-Land-Atmospheric Studies) and the University of California at Berkeley are utilizing a $1.4M award from the National Science Foundation (NSF) to generate new petascale computer models depicting detailed climate dynamics, and building the foundation for the next generation of complex climate models.

For decades researchers assumed that, in some sense, weather and climate were independent. In other words, the large-scale climate determined the environment in which weather events formed, but weather had no impact on climate. However, investigators are finding evidence that weather has a profound impact on climate; a finding that is of paramount importance in the drive to improve weather and climate predictions, as well as climate change projections.


More . . .

Andrew Belliner from Grand Valley State University builds a Personal Supercomputer


Andrew Belliner is a senior student at Grand Valley State University in Allendale Michigan. He built a personal supercomputer to investigate GPU Computing as part of his university program. He is a C programmer and ported his work to CUDA using Smith-Waterman as the projects test bed.

Interactive Map of the World's Fastest Supercomputers

While the U.S. has the world's fastest supercomputers, it faces increased pressure from countries like India and China.

Argonne's Leadership Computing Facility wins the High Performance Computing Challenge


The U.S. Department of Energy's (DOE) Argonne National Laboratory has been named a winner of the annual High Performance Computing (HPC) Challenge Award at the SuperComputing 08 Conference in Austin, Texas.

"It is an honor to be recognized as a winner of the HPC Challenge," said Pete Beckman, director of Argonne's Leadership Computing Facility (ALCF). "This award proves that energy efficiency and computational power are not mutually exclusive. We can still push performance boundaries and deliver stellar results while using a fraction of the power typically needed for supercomputers."

Argonne was the clear winner in two of the four categories awarded in the HPC Challenge best performance benchmark competition, which were run using 32 racks of Argonne's Blue Gene/P.


More . . .

World's Most Efficient Supercomputer Gets to Work


The Fujitsu FX1 computer was inaugurated on Wednesday by the Japan Aerospace Explorations Agency. It has 3,008 nodes each of which has a 4-core Sparc64 VII microprocessor. The machine has 94 terabytes of memory and a theoretical peak performance of 120 teraflops (a teraflops is a trillion floating point operations per second).

Running the standard Linpack benchmark it achieved a peak performance of 110.6 teraflops, which not only ranks it the most powerful machine in Japan but the most efficient supercomputer in the world. Its peak performance represents 91.2 percent of its theoretical performance and outranks the previous record holder, a machine at the Leibniz Rechenzentrum in Munich. Ranked below the German computer is another JAXA machine.


More . . .

Supercomputer – Just how fast?


According to Top500.org, which tracks the world's largest supercomputers, it has taken just over 20 years for supercomputers to increase performance from Gigaflops to Teraflops to Petaflops. In 1986 Cray2 exceeded the Gigaflops barrier for the first time. Eleven years later the barrier was raised to Teraflops by Intel's ASCI Red supercomputer in 1997. IBM's Roadrunner eventually crossed the Petaflops barrier for the first time in 2008, another 11 years after the Teraflops barrier was broken.

The next leap to the Exaflop range is predicted for as early as 2019.


Trends


Hitting these new highs will require significant increases in the numbers of processors underlying supercomputers. Top500.org says that performance improvements will not be achieved by simply increasing per-processor speeds and performance. Instead there will be increasing reliance on clustered multi-core systems to produce the computing power required.

"Going from Terascale to Petascale HPC systems and beyond means that the number of components (cores, interconnect, storage) within such a system will grow enormously. In the near future we will see clustered multi-core systems with core numbers in the range of one hundred thousand to one million and more," the organisation says.


More . . .

NVIDIA Shifts GPU Clusters Into Second Gear


GPU-accelerated clusters are moving quickly from the "kick the tires" stage into production systems, and NVIDIA has positioned itself as the principal driver for this emerging high performance computing segment.

Up until now, all of these GPU-accelerated clusters had to be custom-built. In an effort to get a more "out of box" experience for GPU cluster users, NVIDIA has launched its "Tesla GPU Preconfigured Cluster" strategy. Essentially, it's a set of guidelines for OEMs and system builders for NVIDIA-accelerated clusters, the idea being to make GPU clusters as easy to order and install as their CPU-only counterparts. It's basically a parallel strategy to NVIDIA's personal supercomputer workstation program, which the company rolled out in November 2008.

The guidelines consist of a set of hardware and software specs that define a basic GPU cluster configuration. In a nutshell, each cluster has a CPU head node that runs the cluster management software, an InfiniBand switch for node-to-node communication, and four or more GPU-accelerated compute nodes. Each compute node has a CPU server hooked up to a Tesla S1070 via PCI Express.


More . . .

Top 500 supers - world yawns at petaflops


The annual International Supercomputing Conference kicked off this morning in Hamburg, Germany, with the announcement of the 33rd edition of the Top 500 supercomputer rankings. While petaflops-scale machines are far from normal, they soon will be.

With the June 2009 ranking, the home team in Germany - which has two monster machines in the top ten this time around - will be celebrating. Well, as much as supercomputer nerds celebrate. (We know you are really using the new Jugene and Juropa supers to play video games, at least when the administrators aren't looking. Let's hope the game is not global thermonuclear war).

The Forschungszentrum Juelich (FZJ) has been on a buying binge this year, upgrading its two supercomputers so it can lay the claim of being the floppiest supercomputer center in Europe. The Jugene BlueGene/P system that FZJ bought from IBM packs together 294,912 PowerPC 450 cores running at 3.4 GHz, using a proprietary BlueGene interconnect to deliver 825.5 teraflops of oomph for various research projects, giving it the number three position on the Top 500 list. It runs SUSE Linux - as if you expect anything else.


More . . .

Supercomputers Accelerating Innovation


While the world's challenges seem to be on fast forward, solutions too often unfold in slow motion. Supercomputing is the time machine accelerating innovation to address the biggest scientific challenges.

Oak Ridge goes gaga for Nvidia GPUs


Oak Ridge National Laboratories may not be the first customer that Nvidia will have for its new "Fermi" graphics processor, which was announced yesterday, but it will very likely be one of the largest customers.

Oak Ridge, one of the giant supercomputing centers managed and funded by the US Department of Energy to do all kinds of simulations and supercomputing design research, has committed to using the GPU co-processor variants of the Fermi chips, the kickers to the current Tesla GPU co-processors, in a future hybrid system that would have ten times the floating point oomph of the fastest supercomputer installed today.

Depending on the tests you want to use, the most powerful HPC box in the world is either the Roadrunner hybrid Opteron-Cell massively parallel custom blade box made by IBM for Los Alamos National Laboratory, or the Jaguar massively parallel XT5 machine at Oak Ridge, which uses only the Opterons to do calculations.



More . . .

Large Hadron Collider team flicks switch on Xeon grid


Sverre Jarp, CTO at CERN OpenLab supporting the Large Hadron Collider (LHC) buried beneath the Franco-Swiss border outside Geneva, described the network, powered by Intel Xeons, as "the largest grid service in the world".

Jarp certainly has a big job on his hands. The LHC is to resume operations "within days" following its widely-publicised technical mishaps last year. Once in operation, the mighty proton-punisher will produce colossal amounts of data to be processed.

Briefing reporters at the accelerator facility today, Jarp says that the entire grid, distributed around the world, musters 160,000 cores and measures its storage in tens of petabytes.

"And we are only at the beginning of the demand," he said. "We expect to move into exabytes as the project goes on."


More . . .

Supercomputing 2009 Calit2 Building Collaboration Demo

David Scott (Intel) giving a presentation at SC09 on what's new inside Intel and HPC.

Part 2 of 2. David Scott (Intel) giving a presentation at SC09 on what's new inside Intel and HPC.

A team of IBM members discuss what IBM is showing this week (November 2009) at SC09. Included in the video are Herb Schultz, Matthew Drahzal, Vallard Benincosa, Keith Olsen, Ed Seminaro, and Jimmy Mills.

In this video, John Gustafson (Intel) gives an excellent presentation on what some of the future considerations should be when looking at the next generations of High Performance Computing solutions. I think you will find this very interesting and insightful.

Part 2 of 2.

Wanted: Moore's Law for Another 40 Years


When you fight the law, you'll usually lose... at least that's what happens when you fight Moore's Law, an industry axiom that states that the number of transistors on a chip will roughly double approximately every two years. This video, set to the tune of "I Fought the Law" originally recorded by Sonny Curtis and The Crickets, looks back at how many have long predicted the end to Moore's Law and how it continues to prevail as Intel releases its first 45nm chips based on its reinvented transistors with new Hafnium-based high-k and metal materials. Long live Moore's Law!

Climate modeling requires massive computational power. Until recently, that power required room sized machines with daunting technical and logistic requirements. But new advances in computer design, including hardware and software, continue to facilitate a paradigm shift.

In an effort to broaden and democratize climate research tools, NASA has begun to facilitate the operation of new desktop sized supercomputers, with the goal of making it substantially easier for more researchers to do meaningful work on vital and essential questions for our world.

InfiniBand rockets up Top 500 supercomputer list


InfiniBand-based clusters are charging up the Top 500 supercomputer list with 182 systems, including 63 of the top 100 and five of the top 10 now based on the high-speed interconnect.

Gigabit Ethernet still dominates the list of the world's 500 fastest supercomputers, with 258 machines. But officials with the InfiniBand Trade Association (IBTA) boasted of their momentum this week at the SC09 supercomputing conference in Portland, Ore.

InfiniBand's presence on the Top 500 list, the latest version of which was announced this week, has grown 28% since November. Just four years ago, only 3% of the Top 500 supercomputers used InfiniBand. The technology's growth has come mainly at the expense of Myrinet, an interconnect designed by Myricom that used to hold a substantial portion of the Top 500.


More. . .

Thank You sid2, your posts are really informative and helpful for me. Here at this moment i would like to share with everyone about the concern who are really good with high performance computing. cloudslam09 They are the one who helped me out with my work. I hope this information might be helpful for others too.

DARPA cuts back on Cray super project


As the market was getting ready to open this morning Cray told Wall Street that the US Defense Advanced Research Projects Agency, which funds all kinds of advanced systems and networks, is scaling back its contract for a future petaflops-scale supercomputing system being designed by Cray under DARPA's High Productivity Computing Systems program.

And that is going to chop $60m out of future revenues.


More . . .

Climate and weather modeling with supercomputers.

Climate modeling requires massive computational power. Until recently, that power required room sized machines with daunting technical and logistic requirements. But new advances in computer design, including hardware and software, continue to facilitate a paradigm shift.

In an effort to broaden and democratize climate research tools, NASA has begun to facilitate the operation of new desktop sized supercomputers, with the goal of making it substantially easier for more researchers to do meaningful work on vital and essential questions for our world.

IBM working on 100 GHz Graphene Transistor


Boffins from IBM have emerged from their smoke filled labs with a graphene transistor which can manage 100 GHz when the wind is behind it and it is going downhill.

This is the fastest transistor ever made and could mean that silicon might be put out to pasture as the number one semiconductor. The best silicon transistors ever have done is 40 GHz, and many boffins realised that it will become difficult to squeeze more speed out of the material without coming up with some new breakthrough.

Big Blue seems to think that graphene will get around the problem. Not only is it much faster, but IBM can use the same silicon fabrication techniques in order to make it. It is still early days yet and it will be some time before it trickles its way down to personal computing.


More . . .

Graphene transistor breaks new record


Physicists in the US have made the fastest graphene transistor ever, with a cut-off frequency of 100 GHz. The device can be further miniaturized and optimized so that it could soon outperform conventional devices made from silicon, says the team. The transistor could find application in microwave communications and imaging systems.

Graphene – a sheet of carbon just one atom thick – shows great promise for use in electronic devices because electrons can move through it at extremely high speeds. This is because they behave like relativistic particles with no rest mass. This, and other unusual physical and mechanical properties, means that the "wonder material" could replace silicon as the electronic material of choice and might be used to make faster transistors than any that exist today.


More . . .