Credit: NASA

Just over 50 years ago, on February 20, US astronaut John Glenn blasted into space in his tiny Friendship 7 capsule. His three quick trips around the Earth made him the first American to orbit the planet.

A team of more than 70 IBMers headed by Arthur Cohen as manager of the IBM Space Computing Center in Washington, D.C., had developed the computing systems to manage the launch, orbit and reentry for NASA’s Mercury program. IBM systems manager Saul Gass watched the launch from a grandstand at Cape Canaveral . “Think about the time, 1962. This had never been done before” says Gass, who is professor emeritus at the University of Maryland. ” There was a man in the loop whose life depended on our calculations. It was a demonstration of real-time computing.”

Beginning in the mid-1940s and continuing after the Glenn flight, IBM’s scientists and engineers have contributed substantially to astronomy and manned space exploration, but, today, they’re entering an exciting new phase of  discovery.  IBM scientists in Zurich, Switzerland, and the Netherlands are working with the Netherlands Institute for Radio Astronomy (ASTRON) to develop a massively powerful computing system for harvesting a huge quantity of data gathered by the international Square Kilometre Array (SKA) radio telescope.

The project demonstrates once again the belief that major advances in human achievement and knowledge come through a combination of big bets and bold scientific inquiry.

The telescope, targeted for completion in 2024, will map more than one billion galaxies, with the ultimate goal of exploring the origins of the universe. “These events are the latest in a very long chain of efforts by mankind over hundreds of years to get a better understanding of the universe,” says Martin Schmatz, IBM’s technical lead on the project. “We’re venturing beyond the ability of people to travel in space.”

When President John F. Kennedy in 1961 announced America’s goal of sending a man to the moon within a decade, the technology did not exist to get the job done. But Kennedy trusted that if America’s scientists were given adequate resources and a compelling goal, they could deliver the moon shot. The same is true of the consortium of 20 nations, including European countries and the United States, which are behind the SKA project. They plan on inventing their way to the edge of the visible universe.

The team from IBM and ASTRON is determined to produce advances in chip and computer systems design that will make it possible to handle immense amounts of information–an exabyte of data every day–in real time.  (That’s 1,073,741,824 gigabytes!) Their project, called DOME, will investigate emerging technologies that might ultimately be used in the SKA computing systems. They’ll discover how to transport the data from a giant cluster of radio antennae to a central location, how to filter and store the data, and how to do all of the processing in an energy efficient way. In addition, they’ll develop a system for making all the components work together most efficiently and effectively.

One of their biggest challenges is devising a suitable microprocessor architecture. Right now, according to  Schmatz, the scientists are exploring the idea of packing thousands of low-frequency (and low-power-consuming) processors in a small space and using liquid cooling micro channels to carry away the heat.

While the joint project is focused on SKA, the partners see tremendous potential to use their advances to solve a wide range of so-called Big Data problems. The domains include health care, traffic and natural resources discovery–all of which call for analyzing a tremendous amount of data in real time. The two organizations are setting up the ASTRON & IBM Center for Exascale Technology, based in Drenthe, the Netherlands. Their goal: to boldly explore the far reaches  not just of space but of data.

Click here to view the embedded video.

Technorati Tags: ASTRON, DOME, IBM, Square Kilometre Array

Jason Hlady leads the World Community Grid team at the University of Saskatchewan

When Jason Hlady sees a computer that is turned on but not being used, just sitting there, idling away, he can’t help but think of the possibilities…

That dormant machine could, at that very moment, be running computations to help cure cancer or fight AIDS. It could be solving algorithms that might lead to clean water solutions, or reduce world hunger, or accelerate any number of other worthy research projects.

Hlady, a high performance computing coordinator at the University of Saskatchewan, wants to cut waste and tap the potential of idle computers across the university. To that end, he is leading the drive to get faculty and staff to connect to the World Community Grid — a global network that pools unused computing power and repurposes it for humanitarian research.

As leader of the university’s World Community Grid team, Hlady encourages colleagues to install software that connects their computers to the grid and runs research computations on the machines when they are on, but idle.

“When a computer sits idle, all that energy is just going up a smokestack,” Hlady said. “By joining the World Community Grid, we’re able to put otherwise wasted computing power to good use, helping solve some of the major problems facing our world today.”

Recruiting computers across campus

Hlady’s efforts began nearly two years ago when the University of Saskatchewan became an official World Community Grid partner, committed to support and contribute computing resources to this cause. To date, the university has donated over 250 compute years — about half a year each day — toward humanitarian research via the grid.

“We’ve really just scratched the surface of what we can contribute,” Hlady said. “Just wait until we start to add large labs and entire departments to the grid.” Hlady hopes to boost student involvement as well.

Eventually, the university’s grid team may engage in friendly team challenges with other teams around the world to see which one returns the most results or generates the most run time in a given time period. “If good natured competition can boost enthusiasm for the grid, I’m all for it,” Hlady said.

High praise for grid partnership

For Hlady, becoming an official World Community Grid partner is a smart move for the university any way you look at it.

“Number one, the research being done on the grid is top quality and involves pursuing lofty goals for the betterment of the world — and we want to support that,” Hlady said. “Plus, the projects are targeted and peer reviewed, which is not the case with all public grids.”

“Second, we get experience and develop expertise with BOINC, the open source framework on which most grid and volunteer computing initiatives are built,” he said.

“Third, we think that it’s important to lead by example and show that publicly funded universities can be responsible with their resources,” Hlady said. “By connecting to the grid, we waste less electricity and maximize use of our computing assets.”

A real sense of satisfaction

Hlady runs the grid software on all seven computers in his home and office (connecting to the grid is strictly voluntary for anyone in the university community). His enthusiasm for the grid derives from a passion for research and science that goes back to grade school.

“There’s a real sense of satisfaction in enabling good research,” Hlady said. “Sometimes it’s very difficult to feel like you can make any sort of difference in the world when we’re facing such huge problems. But by contributing to the grid, you can make a real impact.”

Hlady encourages institutions of every stripe to become a grid partner.

“If you make this resource available and show people how easy it is to set things up, they’ll run with it,” he said. “The World Community Grid is a positive thing to be involved with for any organization. There’s really no downside whatsoever.”

World Community Grid pools the surplus computer processing power of more 1.8 million PCs registered by over 570,000 people in 88 countries to tackle projects that benefit all of humanity, like fighting childhood cancer, developing clean energy solutions or designing better treatments to fight AIDS. Volunteers simply download free, secure software that runs quietly in the background when their computer isn’t in use and crunches numbers for humanitarian research initiatives. In terms of pure processing power, the grid is comparable to one of the world’s top fifteen supercomputers.

To read more World Community Grid Person for a Smarter Planet posts, click here.

Technorati Tags: BOINC, Jason Hlady, University of Saskatchewan, volunteer computing, world community grid

Igor Jurisica, Ph.D, uses the power of World Community Grid to conduct his cancer research.

When Igor Jurisica started doing cancer research 11 years ago, he worked with about a dozen colleagues using a handful of scientific workstations in a small lab in Toronto, Canada.

How times have changed.

Today, Jurisica, a senior scientist at Princess Margaret Hospital, Ontario Cancer Institute, conducts his research with the help of nearly 300,000 people spread across 100 countries running his calculations on over 900,000 devices.

This global “team” comprises volunteers who donate their idle computing time to the World Community Grid, creating a virtual supercomputer devoted strictly to humanitarian research. Jurisica’s project, Help Conquer Cancer, is one of nine initiatives that currently share the grid’s massive computing power — free of charge — to conduct critical scientific inquiry.

Jurisica’s research involves computationally intensive calculations to understand the protein crystallization process in general, with a special focus on the structure and function of cancer-related proteins. With conventional computing resources, this project would take at least 186 years to complete.

By using the grid, Jurisica will be able to finish it in just under four years.

“World Community Grid has not simply sped up this research, it has enabled it,” Jurisica said. “The grid has completely transformed the scope of our work and enabled us to finally address our problem in the correct way, in a realistic time frame.”

Visionary cancer research, viable at last

Help Conquer Cancer uses World Community Grid to analyze protein crystals and help improve researchers’ understanding of cancer biology.

Jurisica is using World Community Grid to analyze and classify 115 million images of more than 12,500 human proteins. Each image, created through a process known as x-ray crystallography, displays unique features that must be carefully annotated.

“No human would be able to go through this number of images in a consistent way and accurately classify them,” Jurisica said. ”The grid is the only environment where we can even attempt to do this kind of comprehensive and systematic analysis.”

The results of this project will expand researchers’ understanding of the crystallization process, protein biochemistry and cancer biology and potentially help determine an individual’s predisposition to certain cancers. It may also help improve therapy planning, treatment prognosis and drug development.

Supercomputing power without supercomputer costs

For institutions using World Community Grid, it’s a dream come true: they get all the computational power of a supercomputer without the prohibitive costs and enormous responsibilities.

“Even if somebody gave me Blue Gene or some other machine that could sustain this kind of computation, it’s not feasible because I wouldn’t have the space, power, cooling capacity or the staff necessary to maintain it,” Jurisica said. World Community Grid’s distributed computing model eliminates the headaches of operating a centralized system and lets Jurisica do what he cares about most: focus on his research.

“With the grid, some machines may go offline, new ones will come online, and our computations might move from machine to machine,” Jurisica said. “But I don’t have to concern myself at all with that level of detail.”

Cycling for a cure

When Jurisica is not conducting research, supervising graduate students, traveling to conferences, lecturing or consulting, he likes to go cycling. Since 2008, he has ridden in the annual Ride to Conquer Cancer — a 2-day, 200-mile bike ride from Toronto to Niagara Falls. His IBM-OCI-Roche Integrative Discovery cycling team has raised over C$250,000 for cancer research to date.

“I signed up for my first ride not because I was regularly bicycling, but because of the cause,” Jurisica said. “But since then I have come to love cycling and I’ve even changed my lifestyle to some degree.”

Each summer, Jurisica participates in a 2-day, 200-mile bike ride to raise funds for cancer research.

With his current grid-based research scheduled to wrap up later this year, Jurisica is looking ahead to the next phase of his project, which he hopes to continue running on the grid.

“As a computer scientist, I have great interest in developing new approaches and resources to do complex data analysis and visualization,” Jurisica said. “But it’s really gratifying when this research eventually leads to a clinical test that can hopefully start changing the lives of actual patients. That’s what this is all about.”

World Community Grid pools the surplus computer processing power of more than 1.8 million PCs registered by over 570,000 people in 88 countries to tackle projects that benefit all of humanity, like fighting childhood cancer, developing clean energy solutions or designing better treatments to fight AIDS. Volunteers simply download free, secure software that runs quietly in the background when their computer isn’t in use and crunches numbers for humanitarian research initiatives. In terms of pure processing power, the grid is comparable to one of the world’s top fifteen supercomputers.

To read more World Community Grid Person for a Smarter Planet posts, click here.

Technorati Tags: cancer research, help conquer cancer, Igor Jurisica, protein crystallization, world community grid, x-ray crystallography

IBM's Watson computing system

When we think of the systems that make up a smarter planet, what typically comes to mind are industries like manufacturing, transportation, energy, or banking.  But there is another ‘industry’ that needs to become smarter.  We might call it the humanitarian industry.  That is, the system that creates a safety net to support society and is made up of philanthropies, social services, education organizations, NGOs and government agencies.

In many ways, this is the most human of all systems.  So it is ironic to consider how Watson, a computing system, could help us solve civic, social and cultural challenges and make smarter humanitarian decisions. But Watson’s deep QA technology presents new possibilities to do just that.  Through private sector collaboration with nonprofits, Watson can become the next innovation to be used as a force for societal good.

IBM has a strong track record in leveraging technology to help solve society’s challenges.  We’ve used grid technology to build World Community Grid, DNA analysis to develop the Genographic Project, and speech recognition technology to create Reading Companion.  What we’ve learned through these and other corporate citizenship efforts is that innovation can lead to outcomes with greater societal impact than simple corporate philanthropy alone (i.e. check writing).  We’ve also learned that getting the most from innovations like Watson requires collaboration and partnerships.

That’s why IBM recently invited about 100 leaders from the humanitarian sector to learn about the technology behind Watson and to discuss how we might work together to apply that technology toward achieving societal good.  Ideas ranged from how Watson could help us cut error rates of emergency systems (like 911 and 311) to how Watson might be used as an aid to transform how teachers teach and children learn.  You can learn more about the event by reading this article in Fast Company.

At the end of the event, we encouraged attendees to share their best ideas about how Watson’s technology might be applied to humanitarian endeavors.  I invite you to do so as well.  Please visit the citizen IBM blog to learn more and share your ideas.

Technorati Tags: citizenIBM, citizenship, humanitarian, nonprofit, philanthropy, Watson

Mira, when she’s installed next year, will be a 10-petaflop computer–meaning she’ll be capable of performing 10 quadrillion calculations a second. Argonne already has a supercomputer, Intrepid, based on an earlier version of IBM’s Blue Gene technology. It’s a half-petaflop machine.  Here’s a comparison that gives you a good idea of what a lot more processing power can do for scientists:  Using the current generation of supercomputers operating worldwide, it takes about two years to run a simulation of how a human heart reacts to a new medicine. A 10-petaflop system would cut the wait time down to two days.

Argonne researcher Larry Curtiss hopes all this added computing horsepower will help him develop new materials that could stretch the miles traveled per charge on an electric car battery up to 500 miles, making EVs practical for many uses. “The new computer will help us create the next generation of batteries and make the United States more competitive,” says Curtiss.

Curtiss leads a research team that specializes in nano-materials science. A project devised by his team is among 16 that will get first crack at Mira.  The team will use a computational method called density functional theory to take on materials science problems in four areas–biomass conversion to fuel, electric energy interfaces, catalysis for discovering new materials and lithium-air batteries for EVs.

A key factor in these kinds of experiments is being able to use enough atoms that scientists get a realistic response from the simulation. Working with catalytic processes, for instance, the team has been able to model reactions involving about 1000 atoms. With Mira, they’ll be able to model reactions involving several thousand atoms.

Those little things add up. Fifteen years ago, researchers at Argonne produced some of the scientific breakthroughs that ultimately made possible GM’s new Chevy Volt. Now they’re working on improving the density of batteries, so cars can travel much further on a charge, and on making batteries less volatile, so they won’t explode. If Curtiss and his colleagues succeed, it won’t just be a victory for the United States; it will be a big deal for the entire planet.

Technorati Tags: Argonne Lab, blue gene, electric vehicles, IBM, Larry Curtiss, Mira

The clue: Two of the greatest human Jeopardy! television game show players, ever.  The correct response:  Ken Jennings. Brad Rutter.

Just as IBM set its sights on defeating a chess Grandmaster with Deep Blue in 1997, the company’s scientists have developed a Natural Language Processing, Question Answer machine, named Watson (after company founder Thomas J. Watson, Sr.), to challenge two of the world’s trivia grand masters, to be aired on U.S. television from February 14-16, 2011.

Click here to view the embedded video.

Win or lose on national television, Watson will answer the immediate questions, “does it answer questions accurately?” and “does it answer questions quickly?” with a resounding “yes.”

Beyond excitement for the match itself, the team of IBM scientists is motivated by the possibilities that Watson’s breakthrough computing capabilities hold for building a smarter planet and helping people in their business tasks and personal lives. Watson’s ability to understand the meaning and context of human language, and rapidly process information to find precise answers to complex questions, holds enormous potential to transform how computers help people accomplish tasks in business and their personal lives.

Watson will enable people to rapidly find specific answers to complex questions. The technology could be applied in areas such as healthcare, for accurately diagnosing patients, to improve online self-service help desks, to provide tourists and citizens with specific information regarding cities, prompt customer support via phone, and much more.

Like Deep Blue, Watson represents a major leap in the capacity of information technology systems to identify patterns, gain critical insight and enhance decision-making despite daunting complexity. But while Deep Blue was an amazing achievement in the application of compute power to a computationally well-defined and well-bounded game, Watson faces a challenge that is open-ended and defies the well-bounded mathematical formulation of a game like Chess. Watson has to operate in the near limitless, ambiguous and highly contextual domain of human language and knowledge.

Watson’s technology furthers IBM’s leadership in analytics solutions, which help organizations use the vast amount of information they collect to improve their business operations and service to their customers. Additionally, Watson harnesses IBM’s commercial POWER7 system, showcasing how IBM workload-optimized systems provide unmatched capabilities for processing thousands of simultaneous tasks at rapid speeds, once the realm of only scientific supercomputers.

Read more about the technology behind Watson at ibmwatson.com.

Technorati Tags: business analytics, finance, grand challenge, healthcare, human versus machine, Jeopardy, natural language, question answer, supercomputers, transportation, Watson

Energy efficiency is quickly becoming one of the most important metrics of supercomputing value. Just a few short years ago high performance computing (HPC) clients were concerned primarily with performance, and the cost of performance. The conversation has shifted dramatically. HPC clients are now equally concerned about power consumption and cooling requirements. For good reason, the cost to power an HPC environment today is nearly as much as the hardware.

Fifty percent of the energy consumed in today’s average data center goes toward cooling the systems and preventing overheating. Overheating, in turn, leads to reduced reliability. In fact, Wu-Chun Feng of Virginia Tech, one of the founders of the Green500 List of energy-efficient supercomputers believes that for every 10 degree Celsius increase in temperature the system failure rate doubles.

Significance of the Green500 List

All of this makes the Green500 List of the world’s most energy efficient supercomputers even more compelling, since it clearly calls out which companies are taking supercomputing innovation to the next level. The Green500 List is published by Green500.org and provides a ranking of the most energy-efficient supercomputers in the world. The most current Green500 list published in July reported that 17 of the 20 most energy efficient supercomputers in the world are from IBM. The number one system on the Green500 list is a three way tie between IBM QPACE systems in Germany based on the IBM PowerXCell 8i processor. These top three IBM systems each produce more than 773 Mflops (millions of floating point operations per second) per watt of energy.

IBM developers made energy efficiency a core design principle for HPC systems long ago. For example, when IBM introduced the BlueGene/P supercomputer, it more than doubled the performance of its BlueGene/L predecessor, but only consumed a little more power. The engineering focus on performance per watt is what makes IBM supercomputers the most efficient and consequently, the most affordable. Clients who need to deploy petascale performance systems can save one million dollars a year in energy costs by using IBM’s BlueGene/P, because it is over 40 percent more energy efficient than comparable supercomputers.

IBM recently delivered a first-of-a-kind water-cooled supercomputer to the Swiss Federal Institute of Technology Zurich (ETH Zurich) that consumes up to 40 percent less energy than a comparable air-cooled machine and decreases the carbon footprint of the system by up to 85 percent. The supercomputer, called Aquasar, is made up of water-cooled IBM BladeCenter servers, achieves a performance of six teraflops and has an energy efficiency of about 450 megaflops per watt.

IBM and Arctur, one of Slovenia’s leading software developers, recently announced an agreement to build one of the most powerful supercomputers in the region. Arctur will allow mid-market companies to lease time on the IBM System x iDataPlex high performance cloud computer and help reduce product development time by up to 75 percent. The iDataPlex platform maximizes performance per watt with innovative cooling techniques such as the IBM Rear Door Heat eXchanger. A single iDataPlex 2U server packs five times the computer power of a typical server while consuming 40 percent less power.

As the HPC community pushes toward exascale computing (the performance of one million trillion calculations per second), the energy efficiency design challenges are much greater than the transition from teraflop to petaflop. In previous supercomputing performance generations, the transition was primarily centered on processor development. If current technology design was deployed in exascale computing, such a system would consume between 100 and 200 megawatts (the equivalent energy of a small power plant) and come with a prohibitive power bill of more than $100 million per year.

Most HPC developers agree that an exascale system should only consume about 20 megawatts, which means getting to a 1000-fold increase in performance with only a 10-fold increase in power consumption. IBM is well on its way in research and development to create the next generation of systems that can achieve an exaflop of performance with acceptable levels of power consumption. The formula is a combination of processors, systems level architecture and software.

The global benefits of supercomputing are extraordinary, from pharmaceutical and genetic research to assessing financial risk with pinpoint accuracy or modeling the effects of climate change over the course of a century. Yet all of this work hinges on the ability to balance supercomputing capability with energy efficiency. Now, more than ever, supercomputing performance and power consumption must be developed and implemented with an equal measure of importance.

Dave Turek is vice president of Deep Computing for IBM.