Cutting Edge

IBM researchers have cooked up a quick medical diagnostic testing system based on a silicon chip that can get by on a small sample and test for multiple diseases.

The breakthrough to be announced Tuesday means that physicians can test a patient immediately following a heart attack to improve survival rates. The test checks for disease markers, proteins that can be detected in blood using "capillary action force." In a nutshell, capillary forces refer to the tendency of a liquid to rise in narrow tubes or be drawn into a small opening.

The IBM Research-Zurich findings will be detailed in the December issue of the Royal Society of Chemistry. (See reprint PDF.)

Read more of "IBM researchers speed up medical diagnostic testing via chip" at ZDNet's Between the Lines.

This illustration shows a strand of DNA traveling through a nanopore. With IBM's approach, some layers periodically stop the DNA strand while another measures its properties to determine its genetic information.

(Credit: IBM)

It took 13 years for researchers to catalog all the information in a human genome the first time. Now IBM believes it can do better--somewhat perversely by equipping a newer genetic sequencing method with brakes.

Big Blue is among those who believe electronics technology can be applied to the task of sequencing a person's genes, thereby bringing genetic testing into the computing era and lowering its cost to something like $100 to $1,000.

IBM is working on prototype DNA-processing electronics that slurps strands of DNA through an extremely small hole called a nanopore, measuring the electrical properties of the chemicals as they go by to determine the genetic information. That technique is used beyond IBM, but what Big Blue researchers have been working on is a way to slow down, an essential step toward improving its precision, said Gustavo Stolovitzky, manager of the IBM Functional Genomics and Systems Biology Group.

IBM Chief Executive Sam Palmisano is scheduled to unveil the project and what the company calls its "DNA transistor" Tuesday in a talk, "IT Innovation in Healthcare," at the Cleveland Clinic, IBM said.

The ultimate goal for such research is affordable genetic sequencing. "It would allow DNA sequences to be more or less routine," Stolovitzky said, forecasting that the technology will arrive in five or ten years.

OK, but why should you care?

"It would enable the possibility of going to the doctor with some infection, and the doctor gets the sequence pretty much on spot of the bacteria affecting the patient or the virus is in the blood," Stolovitzky said.

Or another possibility: knowing patients' specific genotypes could mean doctors would know if they had a negative reaction to some drug. That could mean some drugs useful that today are banned could become useful to a subset of the population.

IBM isn't the only one working on this technology. In addition to various academic efforts, start-up 23andMe offers some genetic analysis today.

The genes of animals and plants are encoded in DNA with just four molecular-scale substances--adenine, thymine, guanine, and cytosine. Their particular order governs not only their the formation of humans and other organisms but also the day-to-day biochemistry that keeps us alive.

IBM's sequencing technique to transcribe this biochemical data has been under way for three years, and it's easier said than done. The company is in the process of creating a new prototype device updated to reflect what IBM learned from an earlier one that didn't work as hoped.

"Translocation control we should have in a year's time more or less," Stolovitzky said, referring to the ability to ease the DNA through the nanopore one pair at a time.

The distance scales alone make the work difficult. Each DNA base is about 5 or 6 angstroms away from its neighbor--about half a billionth of a meter. By comparison, a human hair is colossal, about a ten-thousandth of a meter in diameter. And the DNA strands slip through a nanopore that's 2 to 3 billionths of a meter wide.

One problem with the nanopore approach is that it's hard to distinguish the four substances, called bases, as they slip through the hole. The four bases have overlapping electrical properties, so the more time spent measuring each, the better the accuracy.

IBM's approach uses a flat device about 250 nanometers on a side. It has very thin alternating layers of metal and a material called a dielectric. The nanopore is bored through these layers using an electron beam from a tunneling electron microscope, Stolovitzky said.

On one side of the layer is the DNA, unzipped from its familiar double-helix configuration with two strands of matched bases into a single strand with single bases. The single-strand is important in part because the distance between each pair increases to between 5 and 6 angstroms, making them more manageable than the double strand with bases 3.4 angstroms apart, he said.

The strand is pulled through the nanopore by an electrical field that attracts the negatively charged strand. But in the nanopore, some layers are electrically switched on to fix the strand in place for a tick of an electronic clock while another layer makes its measurement, Stolovitzky said.

Even slowed down, the process is fast compared with humans toiling away with pipettes and polymerase chain reaction equipment in a lab. "We think 1 millisecond should be a reasonable time to measure (a base)," Stolovitzky said. In other words, it would take about a second to perform 1,000 measurements.

The human genome has about 3 billion base pairs, so that's still a lot of time to do a full analysis. But it's sill more complicated because the chromosomes that house the genetic data must be broken up into smaller strands for practical reasons.

But IBM Research is happy to pursue a number of projects that may not pay off immediately, including work touching on nanotechnology, computing, and biotechnology. Whether it'll all come to fruition remains to be seen, but one way or the other, it's likely you'll know your own genetic data within a matter of years.

Don Eigler moved the first individual atom 20 years ago, and shortly afterward, he wrote IBM's name with 35 Xenon atoms.

(Credit: IBM)

When IBM researcher Don Eigler picked up and moved the first individual atom 20 years ago today, he paved the way for what arguably was the smallest publicity stunt ever: Big Blue's logo made from a precise arrangement of 35 Xenon atoms.

But moving tiny atoms had big consequences by making the idea of assembling devices atom by atom very real. And the company has built on that nanotechnology foundation, storing information on specific gold atoms, collecting carbon monoxide molecules into computer logic circuits, and pursuing a vision for vastly more compact computing technology.

Despite the progress, Eigler is cautious about when or even if his ideas for computing will bear fruit.

"We did the introduction, and we're in chapter 1," Eigler said. "This is way off in the future, if it ever comes about. I cannot conceive, under the best circumstances, this is going to happen in 10 or 15 years."

Boggled
Eigler, now an IBM fellow, said he was "boggled" that day he moved his first atom with an IBM device called a scanning tunneling microscope. He programmed the system to make the move, then held his breath while his screen went blank during the actual operation.

"You can't see it while you actually move it. Then you see the picture come in and say, 'Yes, it's there,'" Eigler said. He moved the atom back and forth three times to make sure that it really worked: "For us, that's (a) sort of sacred thing. The key thing and most important thing about science is reproducibility. If you can't reproduce your own result, you might as well forget it. It's as if you'd never done it."

Shortly after that, in November 1989, Eigler arranged the 35 atoms to spell IBM. There was, of course, publicity in it for the company, but Eigler had no complaints. For one thing, it demonstrated that IBM really could control atoms with atomic-scale precision and that its work wasn't just a fluke. For another, Eigler was grateful that IBM let him pursue his work.

"It was more than a publicity stunt. Emotionally, for me, it was much more important. This is going to sound hokey, but it's the truth. IBM picked me up off the scrap heap of science and gave me every opportunity a scientist could hope for to be successful," Eigler said. "As far as I was concerned, it was payback time."

No mass manufacturing
Eigler and colleagues have been working on the technology since, but so far, the benefits have been indirect. That's because moving and studying atoms with a scanning tunneling microscope and its offshoot, the atomic-force microscope, is a far cry from assembling computing devices that operate at much larger scales.

"Being able to put atoms together with atomic-scale precision at a level that allows you to deliver a marketable product is something that is largely hope and vision for our future," Eigler said. "We are not there yet."

There are other directions of nanotechnology research; Eigler called out graphene and topological insulators as possibilities. Eigler, though, remains excited to pursue his own long-term vision for computers that process information without today's reliance on the movement of electrons.

IBM Fellow Don Eigler in his lab.

(Credit: IBM)

Specifically, he's interested in using the quantum mechanical property called spin for computing. The conventional conception for this general idea, called spintronics, uses spin to control the flow of electric current, but Eigler wants to use spin alone.

"My goal is to do everything we need to do for computation--logic, storage, information transport--but without moving electrons around at all," Eigler said.

One advantage of the approach is that it avoids electrical current that produces the waste heat that's a major limiting factor in today's computers. Another is that it can enable three-dimensional computing designs much more densely packed with processing power than today's two-dimensional circuitry etched onto silicon wafers.

Spin engineering
The spin of one atom can affect that of its neighbor. The hard part is arranging atoms in order to harness that effect and perform useful computing operations.

"We have to learn how to engineer things so they work the way we want them to work. If you have two atoms, each has spin, and those spins are coupled together in usually two, three, or even four different ways," he said. "You have to place them in the appropriate relationship with one another."

One milestone toward this goal was work by Gerhard Meyer of IBM's Zurich Research Laboratory and others to store data in the form of electrical charges on individual atoms of gold, Eigler said.

In another, IBM's Christopher Lutz found that he could trigger a "molecule cascade," in which a series of carbon monoxide molecules could transmit information. The metastable molecules could store energy, then release it from one neighbor to another similar to a chain of balanced dominoes falling.

Lutz then found a way to arrange those molecules into basic logical processing units of computers, "and gates" and "or gates" that are foundations of today's computers. It didn't use spin, but it's a step in that direction, Eigler said.

Building blocks
One possible intermediate step between moving single atoms and mass manufacturing is what Eigler calls nano plug-ins. If physicists and engineers could figure out how to construct individual logic gates out of a complicated molecule, IBM chemists might be able to figure out a way to synthesize such units in quantity. Next would come the assembly process of snapping these units together appropriately.

"That strategy for building things that work on a very small scale may well be what we see in the future," Eigler said.

And it may arrive, even if his spin-based computation doesn't. "It may be (used with) regular conventional electronics, (or) with carbon nanotubes or graphene," he said. This brings him to the point about why IBM Research invests in such distantly useful technologies.

"The knowledge we're generating in the process of getting there," Eigler said, "is likely to feed into the industry much sooner than the actual outcome--if we ever get to that outcome."

It always happens when you try to do some good, doesn't it?

You try to help an old lady cross the road, and she looks at you harshly and says, "So I look old to you?" You tell that special someone that you love her, and she tells you that she's actually enjoying the company of your best friend.

Such is the painful, ironic circumstance at the United Kingdom's Met Office. ("Met" is short for "meteorological.")

You see, according to the Daily Mail, the agency's large weather brains decided to invest in an even larger IBM brain in order to accurately predict gaseous happenings of climate change.

I have no reason to believe these miserable wet people are from the Department of Communities and Local Government.

(Credit: CC Stevie-B/Flickr)

This metal mastermind can make a quadrillion calculations in the time in takes you to utter a consonant. In the technical world, this is "petaflop" performance. But it might as well have been a Fosbury Flop to some buzz killers.

For along came some bureaucrats from the Department of Communities and Local Government, declaring that the incredibly intelligent hulk is one of the worst polluters in the United Kingdom. For some reason, they were looking at the emissions in all of the nation's public edifices.

All right, so the climate-calculating colossus produces about 75 percent of its own carbon footprint. I fancy that there are several members of Parliament who may do the same. And the supercomputer really can see into the future, whereas some members may not be able to see much beyond lunch.

The beautifully named Barry Grommett from the Met Office told the Mail: "We would be throwing ourselves back into the Dark Ages of weather forecasting, if we withdrew our reliance on supercomputing. It's as simple as that."

Quite. The U.K. bureaucrats have done so much in their attempt to return to the Dark Ages (before climate change?) that the weather men must make a stand.

IBM scientists have imaged the chemical structure of an individual molecule, increasing the possibility for creating electronic building blocks on the atomic and molecular scale.

By using an atomically sharp metal tip terminated with a carbon monoxide molecule, IBM scientists were able to obtain an image of the inner structure of the molecule. The colored surface represents experimental data. The model below shows the position of the atoms within the molecule.

(Credit: IBM)

Scientists In Zurich, Switzerland, have, for the first time, imaged the "anatomy," or chemical structure, of an individual molecule with "unprecedented" resolution, using noncontact atomic force microscopy (AFM), IBM said Thursday. Resolving individual atoms within a molecule has been a long-standing goal of surface microscopy, according to the computer company, which has a research and development program dating back to 1945.

This research will be essential for building computing elements at the atomic scale that are vastly smaller, faster and more energy-efficient than today's processors and memory devices, IBM said.

The research is reported in the August 28 issue of Science magazine.

Though in recent years progress has been made in research of nanostructures on the atomic scale with AFM, imaging the chemical structure of an entire molecule has never been achieved with atomic resolution, according to IBM.

The atomic force microscopy was done in an ultrahigh vacuum and at very low temperatures (5 Kelvin equals minus 268 degrees Centigrade or minus 451 Fahrenheit) to image the chemical structure of individual pentacene molecules. Pentacene has a crystal structure that gives it properties as an organic semiconductor.

Scientists were able "to look through the electron cloud and see the atomic backbone of an individual molecule for the first time." This is roughly analogous to X-rays that pass through soft tissue to enable clear images of bones, IBM said.

The Science magazine article follows another piece published two months ago in the June 12 issue of the magazine covering the "determination of atomic charge states." The results discussed in both of these articles will "open new possibilities for investigating how charge propagates through molecules or molecular networks," IBM said.

Understanding the charge distribution may lead to building computing elements ... Read More

As chip geometries get infinitesimally small, IBM is looking to DNA to make the manufacture of future chips feasible.

On Monday, IBM researchers and collaborator Paul W.K. Rothemund, of the California Institute of Technology, announced an advancement of a method to arrange DNA origami structures on surfaces compatible with today's semiconductor manufacturing equipment.

Low concentrations of triangular DNA origami bind to wide lines on a lithographically patterned surface.

(Credit: PRNewsFoto/IBM)

"The cost involved in shrinking (chip) features to improve performance is a limiting factor in keeping pace with Moore's Law and a concern across the semiconductor industry," said Spike Narayan, a manager in the Science & Technology division of IBM Research, in a statement.

Moore's Law, named after Intel co-founder Gordon Moore, states that the number of transistors that can be placed on an integrated circuit doubles roughly every two years. For more than four decades, chip manufacturers have been able to consistently shrink chip geometries, allowing Moore's Law to remain on track.

But this may not be sustainable for chips with geometries under 22 nanometers. By 2014, the high cost of semiconductor manufacturing equipment will threaten Moore's Law, "altering the fundamental economics of the industry," according to a report released in June by iSuppli. New chip plants typically cost billions of dollars to build, and the tab goes up as chip circuits get smaller.

Individual triangular DNA origami adhere to a template with properly sized triangular features.

(Credit: PRNewsFoto/IBM)

IBM uses DNA molecules as scaffolding--where millions of carbon nanotubes could be deposited and self-assembled into precise patterns by sticking to the DNA molecules. This approach might provide a way to reach sub-22-nanometer lithography--down to 6 nanometers--more economically, according to a paper to be published in the September issue of Nature Nanotechnology, entitled "Placement and orientation of DNA nanostructures on lithographically patterned surfaces." It was co-authored by IBM and Caltech scientists.

"The utility of this approach lies in the fact that the positioned DNA nanostructures can serve as scaffolds, or miniature circuit boards, for the precise assembly of components, such as carbon nanotubes, nanowires, and nanoparticles," according to IBM. The combination of self-assembly with today's fabrication technology eventually could lead to substantial savings in the most expensive and challenging part of the chipmaking process, IBM said.

The lithographic templates, for chip fabrication, were made by IBM using traditional semiconductor techniques, the same used to make the chips found in today's computers, to etch out patterns.

IBM will buy analytics and information forecaster SPSS for $1.2 billion in cash, the companies said Tuesday.

IBM is paying $50 per share for the publicly traded company, which closed Monday on Nasdaq at $35.09. At 6:45 a.m. PDT, the stock had jumped to $49.16.

Chicago-based SPSS makes predictive-analytics software and solutions. Its products tap into vast amounts of customer information that companies can use to try to stay competitive.

Predictive-analytics software is used to gather opinions from customers, forecast future demand, and package the information into business analytics. By capturing and analyzing trends, the software tries to help companies develop products and services better targeted to their customers.

Big Blue has already tapped into the market for predictive-analytics software with its Information on Demand services and its new Business Analytics and Optimization Consulting operation.

IBM said it believes SPSS will provide new solutions for specific industries, such as customer acquisition for financial service companies, patient care for the health care industry, crime prevention for the public sector, and ideal store location for retailers.

"With this acquisition, we are extending our capabilities around a new level of analytics that not only provides clients with greater insight--but true foresight," said Ambuj Goyal, general manager of IBM's Information Management. "Predictive analytics can help clients move beyond the 'sense and respond' mode, which can leave blind spots for strategic information in today's fast-paced environment--to 'predict and act' for improved business outcomes."

Subject to approval from SPSS investors, the deal is expected to close in the second half of 2009. Following the acquisition, IBM will integrate SPSS into its Information Management software portfolio.

The SPSS buyout is just the latest move in Big Blue's drive to win a greater share of the business analytics market. In May, IBM picked up data analytics firm Exeros.

Big Blue's supercomputers are among the greenest in the world.

An IBM supercomputer won first place in a new list ranking the world's most energy-efficient supercomputers.

The June Green500 list, announced June 30 and published by Green500.org, also showed that 18 of the top 20 greenest supercomputers in the world are made by Big Blue.

The group also said that the average efficiency of the supercomputers rose by 10 percent, even as the aggregate power of the machines on the list increased 15 percent.

A key factor in determining a supercomputer's energy efficiency is the number of operations per watt.

Winning the title as most energy-efficient system was an IBM supercomputer based on an IBM BladeCenter QS22 located in Poland at the Interdisciplinary Center for Mathematical and Computational Modeling at the University of Warsaw. The computer produces more than 536 Mflops (millions of floating point operations per second) per watt of energy.

The world's fastest supercomputer, the IBM supercomputer at Los Alamos National Laboratories, came in fourth for energy efficiency, producing over 444 Mflops per watt of energy.

"Modern supercomputers can no longer focus only on raw performance," said David Turek, vice president of deep computing at IBM. "To be commercially viable these systems most also be energy efficient. IBM has a rich history of innovation that has significantly increased energy efficiency of our systems at all levels of the system that are designed to simultaneously reduce data center costs and energy use."

The Green500 group also noted that the No. 5 supercomputer, GRAPE-DR of the National Astronomical Observatory of Japan, is "arguably" the first on its list with more than a million processing elements--in this case, 2.1 million.

Unveiled in 2007, the Green500 list is published two to three times a year by Green500.org. It typically serves as a follow-up to the Top 500 list of worldwide supercomputers announced by Top500.org. In the most recent Top 500 list revealed last month, the Los Alamos supercomputer built by IBM hit a peak performance of 1.105 petaflop/s (quadrillions of floating point operations per second).

Roadrunner maintains its lead as the fastest supercomputer in the world.

(Credit: IBM)

Despite the Jaguar nipping at its heels, Roadrunner continues to speed past the supercomputing pack.

That's according to the twice yearly Top500 list of the fastest supercomputers in the world, which is to be announced Tuesday morning at the 2009 International Supercomputing Conference in Hamburg, Germany. The list is released in June and November every year.

The IBM supercomputer housed at the Department of Energy's Los Alamos National Laboratory, known as Roadrunner, maintains the lead it grabbed a year ago. The computer can process 1.105 petaflop/s, or quadrillions of floating point operations per second, according to the Top500 Linpack benchmark. Hot on its heels for the second year in a row is the Cray XT5 Jaguar system at the DOE's Oak Ridge National Laboratory, which clocked in at 1.059 petaflop/s.

Despite the consistency of those top two systems, there were some newcomers to the top 10 of the list of 500 this year, and not from within the U.S. The new IBM computer, known as JUGENE, installed at Forschungszentrum Juelich in Germany hit 825.5 teraflop/s, or trillions of floating point operations per second, which was good enough for third place on the list. Forschungszentrum Juelich also is home to the 10th place supercomputer, JUROPA, which is a combination of Bull Novascale and Sun Sunblade x6048 servers. It achieved 274.8 teraflop/s.

The rest of the top 10 fastest computers in the world are all housed in the U.S. But some notable international sites are demanding attention. An IBM BlueGene/P system at King Abdullah University of Science and Technology in Saudi Arabia took 14th place, while the Dawning 5000A at the Shanghai Supercomputer Center in China took 15th place.

The threshold to get on the Top500 list this year got increasingly tough. The slowest computer on the list hit 17.1 teraflop/s, when six months ago the slowest computer on the list achieved 12.64 teraflop/s. That also means the total combined power of the 500 supercomputers is faster than ever at 22.6 petaflop/s. Six months ago the top 500 hit 16.95 petaflop/s, and 11.7 petaflop/s a year ago.

Despite holding some of the top spots, IBM's overall dominance as the top supplier of servers for these supercomputers has been eclipsed by Hewlett-Packard. While IBM leads in overall installed performance, HP has the greater market share at 212 to IBM's 188.

Inside those servers, Intel has the lion's share of processors, with just under 80 percent, or 399 of the top 500. IBM Power processors are the second-most popular, and can be found in 55 of the systems.

The Top 10 List:

• Roadrunner, IBM, Los Alamos National Laboratory (1.105 petaflop/s)
• Jaguar, Cray, Oak Ridge National Laboratory (1.059 petaflop/s)
• JUGENE, IBM, Forschungszentrum Juelich (825.5 teraflop/s)
• Pleiades, SGI, NASA Ames Research Center (487.01 teraflop/s)
• BlueGeneL, IBM, Lawrence Livermore National Laboratory (478.2 teraflop/s)
• Kraken XT5, Cray, National Institute for Computational Sciences (463.3 teraflop/s)
• BlueGene/P, IBM, Argonne National Laboratory (458.61 teraflop/s)
• Ranger, Sun, Texas Advanced Computing Center (433.20 teraflop/s)
• Dawn, IBM, Lawrence Livermore National Laboratory (415.70 teraflop/s)
• JUROPA, Bull SA, Forschungszentrum Juelich (274.80 teraflop/s)