CRL’s EKA Supercomputer
(Photo Credit: TechGadget.in)

My home PC is six years old; to me, these numbers are so big as to have no meaning.  But here are some stats, anyway: "Called the EKA, CRL’s supercomputer . . . has 14,400 processors, 28 terabytes of memory, 140 terabytes of disks, a peak performance of 180 trillion calculations per second (180 teraflops), and sustained computation capacity of 120 teraflops for the LINPACK benchmark."

Sounds like the sort of machine that could find the answer to life, the universe, and everything, eh?

But, at least in the short term, the EKA and this partnership "will help bridge the gap between traditional supercomputing and cloud computing research in India," according to a statement by S. Ramadorai, the chairman of CRL.

More specific goals, dates, and dollar amounts (it’s possible that money changed hands) remain unknown.  Hopefully we’ll hear something within the next 7.5 million years.

Yahoo Queries The Obscure

Such frustration serves no one, not the searcher, and definitely not the search engine as it fails to place relevant advertising alongside the search results a person hopes to see. It’s a situation, as Resource Shelf indicated from Yahoo Research, that has room for improvement.

The fine minds at Yahoo Research, who most recently enjoyed their supercomputer debut at Carnegie Mellon, tackled the problem of obscure queries. Their research yielded a way to improve upon the one-off searches that engines sometimes see:

To address the problem, the Yahoo! team proposed a methodology for using search results, as well as information available on the Web, as a source of external knowledge. To this end, they sent rare queries to a search engine and assumed that a majority of the highest-ranking search results were relevant to the query. Categorizing these results allowed the team to classify the original query with high accuracy

The results definitively confirmed that using the Web as a repository of world knowledge contributes valuable information about the query, and aids in its correct classification. “We discovered the best source of information to understand what these rare queries are about is to look at the search results,” Broder explains. “If you look at each returned page as a vote on what the query is about, you find that the majority tends to be correct even though many individual pages are wrong.”

A couple of positive results should happen as research continues. Search results should be more in line with what the searcher expects, while advertisers with the most appropriate messages to place with those results may find better responses to their campaigns from qualified customers.

follow me on Twitter

For those of you firing up the old faithful laptop for a morning of surfing, blogging, maybe a little development work, get a load of what some of the lucky geeks at Carnegie Mellon University got to play with this morning:

The M45, Yahoo’s supercomputing cluster, has approximately 4,000 processors, three terabytes of memory, 1.5 petabytes of disks, and a peak performance of more than 27 trillion calculations per second (27 teraflops), placing it among the top 50 fastest supercomputers in the world.

Their ranking claim won’t be confirmed until the next Top500 Supercomputer list comes out on Tuesday at this week’s SC07 conference in Reno, so it will be interesting to see how M45 measures against the best in the world. Yahoo’s M45 figures should put it in the top 30.

We chatted with Yahoo’s Ron Brachman, VP for worldwide research operations with the company. He’s also wearing the hat as head of academic relationships. Jay Kistler, VP for engineering system tools & services, also talked with us ahead of this morning’s announcement.

Brachman said the M45 supercomputer came about from the opportunity for Yahoo and the university community to advance science and technology on an Internet scale. They have opted to focus on open source, developing solutions for large scale distributed computing.

Yahoo and Carnegie Mellon understand grid computing well. The M45 setup has been geared toward that understanding. It’s capable of partitioning large data sets thanks to the installation of Hadoop.

Hadoop accomplishes this by implementing MapReduce and Pig the latter which may be known to those who follow Yahoo’s research projects closely.

Kistler said they have been working on layering Pig over a Hadoop core. Pig’s runtime extensions for parallel computing are similar to SQL, but they are procedural rather than declarative.

In the M45 environment, the runtime maps statements down to where MapReduce can divide them into little blocks of work and run them across the supercomputing platform.

We wanted to understand better what the distributed development effort being enabled by M45 might be able to do for this level of supercomputing. Kistler rattled off a couple of achievements he would like to see happen, if developers can pull them off.

One would provide for the improvement of job scheduling across clusters; another the enhancement of monitoring and instrumentation of heterogeneous jobs, where it would be easier to find bottlenecks and faults, and correct them for better performance.

Compelling stuff for the folks who will really get into the tasty innards of supercomputing. The potential gains from the M45 go beyond the items on Kistler’s wish list.

Carnegie Mellon’s Randy Bryant, dean of the School of Computer Science, told us in a phone interview about such possibilities. Top of the list: generating statistics for language translation. It’s a demanding task due to the number of documents needed for mapping words from multiple languages.

Another potential gain would be with digital image editing. Bryant discussed this with the example of getting an ex-brother in law out of photos. Through the use of a massive digital image database, supercomputing could allow the editor to find the content of a photo minus the person to be edited out, and replace that person with the background that would be normally visible.

Semantics and language search support would benefit, and we think Yahoo will be interested in that. Bryant noted such a project would look at distinguishing linguistics, where the system would understand when a speaker means "bare" or "bear" from the context of the rest of a conversation.

Research takes time, but the M45 platform should substantially improve the total time needed for these projects to bear productive results. Some very lucky geek types started researching on this platform today.

follow me on Twitter

At Virginia Tech, the System X supercomputer consists of a cluster of 1,100 of those Apple Xserve G5 machines.

The PhysOrg website reported on how researchers at the school have developed a method of finding answers by combing through seemingly disparate events.

Those researchers call the search capability they’ve created, “Storytelling.” It is a lot more than keyword search, and sounds more like the conceptual search IBM has been working with its UIMA project.

The Storytelling process has been described as finding a “chain of concepts” between specified start and end points.

A researcher on the project described the process:

It’s pretty amazing technology at work, and it sounds like they can find answers to questions that are a little deeper than my last web search for Midnight Oil lyrics.

—

Add to Del.icio.us | Digg | Yahoo! My Web | Furl

Bookmark WebProNews:

David Utter is a staff writer for WebProNews covering technology and business.

Currently, IBM’s already mind-blowingly fast Blue Gene supercomputer is capable of 136.8 teraflops, or 136.8 trillion calculations per second.

But Japan thinks it can do 73 times better by building one operating at 10 petaflops, or 10 quadrillion calculations per second.

The five-year project is expected to cost between 80 and 100 billion yen ($714 million to $893 million).

The ever-careful decision making Japanese government said the plans are not yet approved but a decision will be made by the end of August. If approved in the national budget, the supercomputer is expected to be delivered by March 2011.

Still wounded after the Blue Gene took over the world’s fastest computer over Japan’s Earth Simulator supercomputer in 2004 (operating at a crawling 35.9 teraflops), the Land of the Rising Sun hopes to reclaim the crown.

While the Earth Simulator tracks global sea temperatures, rainfall, and a number of geological indicators to predict natural disasters, the next generation of supercomputer is hoped to expand the capabilities to simulating galaxy formation and human body reaction to various medicines.

The Department has designated the Aeronautical Systems Center (ASC) Major Shared Resource Center (MSRC) at Wright-Patterson Air Force Base in Ohio, according to a press release.

The cluster will be capable of 10 teraflops, or 10 trillion floating-point operations per second. In a win for AMD, its Opteron processors will be on-board the HP ProLiant 145 servers delivered to the ASC MSRC.

“This deployment further illustrates the market’s acceptance of cluster technologies for supercomputing challenges in real world deployments,” said Winston Prather, vice president and general manager, High Performance Computing Division, HP.

HP also has orders for a smaller 8-node development system at Wright-Patterson, and a 46-node system for Arnold Air Force Base in Tullahoma, Tennessee. At ASC MSRC, the HP machines will join a variety of SGI, Sun, and Compaq hardware already in place.

The ASC MSRC provides high performance computing resources to a variety of Defense projects. They emphasize particular computational tasks, like fluid dynamics, structural mechanics, and nano-electronics.

David Utter is a staff writer for WebProNews covering technology and business. Email him here.

The world’s leading authority on supercomputer ranking, TOP500, has announced that the IBM BlueGene/L supercomputer is the most powerful one in the world, for the second consecutive year.

IBM’s newest supercomputer, Watson Blue Gene based in Yorktown, NY, came in second on the list, where IBM took six of the top ten spots. WBG ranks as the most powerful privately owned supercomputer today.

BlueGene/L was measured as having a top performance of 136.8 teraflops, which would be trillions of floating point calculations per second. Through the summer, IBM expects that performance to increase to 360 teraflops.

With 259 systems on the list, IBM becomes the first vendor to have more than 51 percent of the total number of ranked systems. “The latest list, particularly if you look at the Top 10, clearly illustrates the dynamic nature of supercomputing today,” said Erich Strohmaier, one of the founding editors of the TOP500 list.

BlueGene/L performs work for the government on one of the most critical areas of research in the world. “We are doing science critical to (the) National Nuclear Security Administration’s mission to ensure the safety, security and reliability of the nation’s nuclear weapons stockpile.

“This represents a great team effort led by NNSA’s Advanced Simulation and Computing program,” said Dona Crawford, associate director for Computation at Lawrence Livermore National Laboratory.

Out of the 500 listed machines, 304 were Linux commodity clusters. Eight of the top ten supercomputers run Linux. Linux has grown from a dorm room creation to being a dominant presence on the world’s most powerful computing hardware.

David Utter is a staff writer for WebProNews covering technology and business. Email him here.

It consists of twenty refrigerator-sized racks and boasts a processing speed of 91.29 teraflops. As it comes online, BGW will be one of the top three supercomputers in the world, according to an IBM press release.

IBM scientists will explore how BGW might help progress in many fields. While business applications will be among these, they anticipate gains in fields like life sciences, quantum chemistry, and fluid dynamics.

One of the first applications to be deployed on BGW will be Blue Matter. Initial results on Blue Matter, which is used to run protein dynamics simulations important to drug development, were published in the Journal of the American Chemical Society in April 2005.

Further, IBM intends to actively seek academic and industrial research, as part of the Department of Energy’s Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program.

The program seeks computationally intensive, large-scale research projects and encourages proposals from universities, other research institutions and industry.

BGW will also be used by IBM’s Center for Business Optimization (CBO) — a new consulting and software unit that taps IBM’s math scientists, industry and deep computing expertise to tackle clients’ previously unsolvable problems.

For example, high-precision weather forecasting software will feed predictive models for such diverse applications as disaster response, utility supply/demand forecasting, agricultural maintenance scheduling and transportation planning.

IBM anticipates that Blue Gene’s combination of high performance with smaller size, cost and power consumption has brought supercomputing technology to the point where it can now be made more widely available.

No word yet on whether BGW runs Solitaire, though.

David Utter is a staff writer for WebProNews covering technology and business. Email him here.

BusinessWeek Online indicates Blue Gene’s new record may not stand for very long either. Blue Gene is expected to have the hardware producing these extravagant number to double, meaning 184 teraflops will more than likely double to an eye-popping 368 teraflops per second.

Blue Gene was designed for the National Nuclear Security Administration in order to simulate the performance and safety of nuclear weapons and other applications.

The BusinessWeek Online article also provides a top-25 list of the fastest supercomputers in the world. An image of the list follows:

Click For Larger Image

Chris Richardson is a search engine writer and editor for WebProNews. Visit WebProNews for the latest search news.

One week after a similar deal with the Arctic Region Supercomputing Center, Big Blue said the Texas Advanced Computing Center (TACC) had installed 32 super-dense cluster of IBM eServer p655 servers and one p690 system.

Click below to read this article at the Internetnews.com website:

http://www.internetnews.com

Internetnews.com contains award-winning real-time news focused 100% purely on the Internet industry. Internetnews.com is part of the internet.com Network of websites and email newsletters which are owned by Jupitermedia. Jupitermedia provides business professionals with the up-to-the-minute technology news, resources and product information that they need to do their jobs. Our internet.com and EarthWeb.com network consists of over 150 targeted Web sites and 200 e-mail newsletters serving vertical markets in the technology sector and serve over 20 million unique users each month.