Accelerator physicists strive to lower cost of cancer treatment

May 28, 2010 | 6:56 am

HIMAC in Chiba, Japan, was the first carbon-ion therapy center to take patients, in 1994. The accelerator sends carbon ions into three treatment rooms. Image courtesy of NIRS.

Several facilities that will offer cancer patients the latest innovation in hadron therapy, a medical application of particle accelerators, are under construction in Europe and Asia.

But so far the high cost of building and operating these facilities has prevented the treatment from becoming widely available.

Accelerator physicists from industry and academia challenged one another this week at the International Particle Accelerators Conference in Kyoto, Japan, to find ways to make the treatment, carbon-ion therapy, more affordable.

Hadron therapy uses an accelerator to send particles such as protons or ions into a patient’s tumor. The particles travel through the patient’s body and release most of their energy into the tumor cells, damaging them while limiting harm to the surrounding healthy tissue.

The development of proton therapy was a huge advancement in the treatment of cancer and is often just as effective as the more expensive carbon-ion therapy, said William Chu of Lawrence Berkeley National Laboratory in a talk on the subject of carbon-ion therapy this week at IPAC ’10. It has been the most widely used form of the treatment.

By the end of 2009, about 78,000 patients worldwide had been treated using hadron therapy, according to the Particle Therapy Co-Operative Group, PTCOG. About 86 percent were treated with protons, and less than 10 percent–about 7000 patients–with carbon ions.

Both treatments damage the DNA in a tumor cell, which can lead to its death, said Chu, a retired physicist who worked on hadron therapy at Lawrence Berkeley National Laboratory between 1975 and 1993.

DNA is made up of two strands of units called nucleotides that form a double helix. Each nucleotide in a strand has a partner nucleotide in the other strand. Proton therapy usually destroys a nucleotide in a single strand, leaving its partner nucleotide intact. The partner nucleotide can tell an enzyme how to replace the nucleotide that was destroyed. This way, the body can repair the damaged cancer cell.

Carbon ions, on the other hand, are more likely to damage both strands of DNA. This leaves enzymes without instructions on how to repair the cell and makes it more likely to die.

Using carbon ions also can be more effective in tumors with large centers void of the dissolved oxygen that blood vessels deliver, Chu said.

Four facilities currently offer carbon-ion therapy: the Heavy-Ion Medical Accelerator in Chiba, Japan; the Hyogo Ion Beam Medical Center in Hyogo, Japan; GSI in Darmstadt, Germany; and the Heidelberg Ion Beam Therapy Center in Heidelberg, Germany. Six new carbon-therapy facilities are under construction in Wiener Neustadt, Austria; Pavia, Italy; Heidelberg, Marburg, and Kiel, Germany; and Maebashi, Japan.

However, Chu said, building and operating a facility for carbon-ion therapy costs about twice as much as building and operating one for proton therapy–already an expensive venture at $120 to $180 million. So accelerator physicists will need to develop cheaper ways to offer the treatment if it is to gain the prominence of its more popular relative.

Kathryn Grim

1 Comment »

Middle East accelerator project approaches barrier

May 27, 2010 | 10:56 am

The BESSY I synchrotron-light source was originally operated at the Berlin Electron Storage Ring Company for Synchrotron Radiation. Image courtesy of BESSY.

Members of an unlikely international collaboration constructing the Middle East’s first synchrotron light source have dealt with outdated equipment, inexperience, and language barriers.

But one hurdle looms particularly large in their path: They need to gather more than $24 million to complete the final section of the accelerator.

The collaboration has struggled to find the funds within its membership and has begun discussions in Europe and the United States, said technical director Amor Nadji in a talk at IPAC ’10.

A synchrotron accelerator uses magnets to circulate electrons at almost the speed of light, creating a beam of bright ultraviolet and X-ray light. Scientists use beams from synchrotrons in materials science and biomedical applications. For example, biologists used a synchrotron light source to establish the double-helical structure of DNA.

The Synchrotron Light for Experimental Science and Applications in the Middle East, or SESAME, represents a rare example of cooperation between Bahrain, Cyprus, Egypt, Iran, Israel, Jordan, Pakistan, Palestinian Authority, and Turkey.

Young scientists from around the contentious region have been working side-by-side on construction of SESAME since 2003.

“The purpose is to try to use science for peace,” Nadji said.

Physicists Herman Winick, Gustav-Adolf “Gus” Voss, and others planted the seed for the project in 1997 when they proposed that Germany donate rather than scrap a recently decommissioned synchrotron called BESSY I.

As scientists in Germany began construction of the synchrotron’s successor, BESSY II, others packed the components of the original into crates and shipped them to a town near Amman, Jordan.

The process of fitting the pieces back together into a working accelerator has not been easy. Scientists have had to find replacements for parts built in the ‘70s and decipher notes handwritten in German. On top of that, most of those willing to work on SESAME are fresh out of school.

“Building an accelerator is not something you can learn in class,” Nadji said. “You need to go to the laboratory.”

More experienced scientists tend to go abroad to participate in projects with more stable funding, which offer higher salaries. Nadji grew up in Algeria but moved to France to complete his PhD. He worked at the Laboratoire pour l’Utilisation du Rayonnement Electromagnetique, or LURE, and more recently helped build the Soleil synchrotron near Paris.

SESAME is located near Amman, Jordan.

Nadji said he enjoys the teaching aspect of his job at SESAME.

“These are inexperienced people,” he said. “But if they have confidence in you, it’s a very good atmosphere. Every day at the end of the day, I think I did something good because I told this one about a strategy he didn’t know or because I know that one can manage something on his own now.”

Despite their freshness, members of the SESAME collaboration are not content simply to rebuild the German synchrotron. They plan to upgrade the accelerator to run at 2.5 gigaelectronvolts, up from its original 0.8 GeV.

The improvement necessitates a storage ring about twice the size of the original and raises the cost of building the accelerator from its original estimate. But it will make SESAME competitive with other synchrotrons around the world, an important factor in attracting veteran scientists to use the machine.

Read more about the origins of SESAME in symmetry.

Kathryn Grim

1 Comment »

Physicists hold first international particle accelerator conference

May 24, 2010 | 10:27 am

For decades, scientists from laboratories and universities around the world have worked together to build and operate particle detectors.

Until recently, accelerator physicists have done just the opposite, working predominantly with fellow scientists at the same institution to build their machines.

This week hundreds of accelerator physicists have gathered in Kyoto, Japan, to take part in the first International Particle Accelerator Conference, taking a step toward the practices of their detector-building colleagues. “Any new accelerator will surely need to be built in an international collaboration,” said Katsunobu Oide, chair of the IPAC 2010 organizing committee, in his introductory remarks.

Detector builders expect to work with collaborators from outside their institutions. If a particle physicist wants to take part in an experiment, he or she must take some responsibility for the detector by helping with construction or taking shifts running it.

There is no similar expectation that the scientist help build or run the accelerator that delivers particles to the detector. The physicists, engineers, and technicians that do this historically have had little assistance from outside the host institution.

One reason for this is that operating an accelerator is more complicated than operating a detector, Oide said.

“A detector is more of a passive device,” he said. “Once you build it, you use it.” The major work takes place during construction.

But an accelerator requires constant, specialized attention. “It cannot be automatic if you want cutting-edge performance,” he said.

Those who run particle accelerators need to work in close proximity to the accelerator so that they can investigate if something goes wrong. The philosophy has been that if locals are going to operate the accelerator, they might as well build it, too.

But particle accelerators have grown larger and more complex over time. It is no longer feasible for a single institution to provide the manpower and funding necessary to complete construction.

The Hadron Electron Ring Accelerator, or HERA, completed in 1992 at DESY in Hamburg, Germany, was the first accelerator to receive a significant in-kind contribution from a collaborating institution. Italian scientists built half of its 416 superconducting dipole magnets.

Since then, multiple institutions have worked together on accelerators such as the Large Hadron Collider and Europe’s XFEL. Proposed future accelerators such as the International Linear Collider, the Compact Linear Collider, and the muon factory are set to follow this trend.

This week, accelerator scientists at IPAC 2010 hope to find new ways to capitalize on the advantages and manage the disadvantages that the new experience of working as a team can bring.

Kathryn Grim

No Comments »

Visiting a physics conference in Japan

September 11, 2009 | 11:02 am

An international group of physicists-many of them young-met in Kobe, Japan, last week to discuss the experiments and theories of the energy frontier.

The annual Physics in Collision conference began in the United States in Blacksburg, Va., in 1981. For 29 years, the conference has given particle physicists a chance to present their work and discuss the with one another what it means.

In this video, one of the hosts of PIC 2009 and a future host of PIC 2010 offer a flavor of what an international physics conference is all about.

Kathryn Grim

No Comments »

The PAMELA spacecraft's view of the Van Allen radiation belts

September 4, 2009 | 7:15 am

Two rings of highly energetic charged particles encircle the earth, trapped in its magnetic field. The inner ring, made up mostly of protons, hovers about 500 miles above our heads. The outer ring is mostly made up of electrons.

These radiation belts are named for astrophysicist James Van Allen, who first discovered them. Though the Van Allen belts are invisible to the naked eye, an instrument called PAMELA gives physicists an idea what they would look like.

The Payload for Antimatter Matter Exploration and Light-nuclei Astrophysics, or PAMELA, telescope was launched in June 2006 from Kazakhstan in part to study the radiation belts. Its goals are to search for dark matter, baryon asymmetry, and the source of cosmic rays. Marco Casolino, a researcher at the National Institute of Nuclear Physics in Rome, described the instrument in a talk at the Physics in Collision conference in Kobe, Japan, this week.

To illustrate its capabilities, he showed the following fascinating animation:

PAMELA’s view of the Van Allen belts

The video gives a tour of the Van Allen belts, as measured with PAMELA spectrometer on board Russian satellite Resurs-DK1. Measurements are between 600 and 370 km and are shown to scale.

Kathryn Grim

4 Comments »

Top quark chefs

September 2, 2009 | 11:52 am

This collision event display, created by the CDF collaboration, shows a single top quark candidate event. Credit: CDF collaboration.

The top quark is the heaviest fundamental particle.

Fundamental particles are the most basic building blocks of matter; they cannot be broken into parts. Even though it is so small, the top quark is as heavy as a gold atom, which is made up in part of almost 200 protons and neutrons.

“We think the top quark might be special because it is so massive,” said Bernd Stelzer, an experimental physicist on the CDF collaboration. “We want to look at it from all angles.”

Fermilab physicists are examining the production, properties, and decay of top quarks to gain the most complete picture of the particle possible. They compare their observations to predictions made in the Standard Model of physics and in theories that build on that model.

The CDF and DZero collaborations at Fermilab released 25 new experimental results this summer that pushed top quark measurements to higher and higher precision, Stelzer reported at the Physics in Collision conference in Kobe, Japan, this week.

The tiny particle with a huge mass has inspired and factored into many theories, some of them wild.

In one theory, the three-dimensional world we experience is only one layer of a four-dimensional universe. The closer a particle that lives in the warped fourth dimension comes to our slice of life, the closer it is to the generation of mass. So, according to the theory, the top quark is so heavy because it is the closest fundamental particle.

Other physicists have suggested that the Higgs, the theoretical particle that gives other particles mass, is actually made up of pairs of heavy top quarks bound together.

Fermilab physicists recently made strides toward understanding the curious particle.

Fourteen years ago, the CDF and DZero collaborations discovered the top quark produced in particle-antiparticle pairs in Fermi National Accelerator Laboratory’s particle accelerator, the Tevatron. In March, CDF and DZero scientists announced the first observation of particle collisions resulting in single top quarks.

“The new single top quark sample can serve as a new compass to point us to new physics,” Stelzer said.

A proton-antiproton collision can produce a single top quark in two different ways. Credit: DZero collaboration.

The collisions that create top quark pairs always behave the same way, but single top quarks are created in two distinct ways. The Standard Model predicts how often each of the two processes will create a single top quark, as do the many exotic theories about the tiny, tubby particle. Measuring the rates of these processes precisely will allow physicists to determine which theories came up with the right prediction.

A virtual helping hand

Collisions of protons and antiprotons in the Tevatron convert energy into mass, as described by Einstein’s famous equation, E=mc². Accelerator operators accelerate particles to high energies and smash them together, which creates a spray of short-lived particles with a mass that corresponds to that energy.

When the collision energy of the machine corresponds to the mass of a particle, the rate of producing that particle in the collisions shoots up rapidly. This is called resonance production.

At an energy of about 80 GeV, the Tevatron produces a plethora of W bosons. When turned up to about 91 GeV, it begins to create Z bosons.

But there is no such recipe for top quarks.

The collisions that create both pairs of top quarks and single tops borrow energy from virtual particles. Virtual particles wink in and out of existence for only brief moments, but they affect interactions between particles around them.

The more massive a particle, the more infrequently it appears as a virtual particle. But even very massive particles can appear this way. That’s important, because an interaction needs a very large virtual particle to create a heavy top quark.

Physicists cannot predict the emergence of a virtual particle by the energy of the machine. Top quarks appear to lack resonance production.

At the conference, Stelzer also discussed efforts at the Tevatron to find evidence for resonance production of top quarks. This would point to top quarks being created in a new way not predicted by the Standard Model, without the help of virtual particles.

“That would be fantastic,” Stelzer said. “It would be an unambiguous sign of new physics and would likely allow us to reveal the origin of the enormous mass of the top quark.”

If CDF and DZero do not find this evidence of new physics, researchers will continue to search for it at experiments at the Large Hadron Collider at CERN.

Kathryn Grim

1 Comment »

Building a mystery with the g-2 experiment

September 1, 2009 | 10:28 am

Experimental physicists love a mystery. When an experiment does not turn out the way theorists predict, it can mean the discovery of new physics.

So it might seem contradictory that one thing a physicist simply cannot tolerate is uncertainty.

Experiments studying subatomic particles called muons recently illustrated the difference between the two and spurred experimentalists to call for further study.

The trouble with uncertainty

g-2 experiment at Brookhaven National Laboratory / BNL

Muons are the heavy cousins of electrons. Both muons and electrons have a property called a magnetic moment, which determines how they act in a magnetic field. The equation to calculate the magnetic moment of a muon depends on a number represented by the letter g, which early theory had predicted was equal to 2. Subsequent experiments found a different value for g.

In 2004, an experiment called g-2 (g minus two) at Brookhaven National Laboratory on Long Island in New York made the most precise measurement of g-2 until that point.

But the results were uncertain. The theoretical prediction had an amount of uncertainty. The experimental result had an amount of uncertainty. Both amounts were smaller than one part in a million.

The difference between the two numbers and the combined uncertainty just barely left scientists without a clear answer as to whether they had actually discovered a difference between theory and experiment.

“The goal of the experimentalist is to be able to say something meaningful, so we hate these small-significance deviations,” said Denis Bernard, experimental physicist at Ecole Polytechnique in France. “Each time we have a small-significance deviation, we work to improve the precision and eliminate a mere statistical fluctuation–or better, discover the next big thing.”

In a lecture at the Physics in Collision conference in Kobe, Japan, this week, Bernard presented the results of a new method of determining the theoretical measure of g-2.

Physicists in the BaBar collaboration at SLAC National Accelerator Laboratory submitted the results of the new measurement to Physical Review Letters last week.

The new measurement was more precise, and the result brought the theoretical prediction for g-2 and its experimental measurement closer together.

However, the significance of the difference between the two numbers remains maddeningly the same.

It works like a poll. If Candidate A is predicted to earn 60 percent of the vote and Candidate B is predicted to earn 40 percent, but the margin of error is 11 percentage points, the poll does not actually predict a winner.

But if the two percentages are closer together–say, 51 percent to 49 percent–the margin of error can be lower–say, 2 percentage points–and still render the prediction of who will win just as meaningless.

Bernard said a new g-2 experiment, which some physicists hope might take place at Fermi National Accelerator Laboratory in Illinois or at High Energy Accelerator Research oragnization (KEK) laboratory in Japan, could finally put an end to the frustration.

Either the theory and experimental results correspond or the numbers are far enough apart that scientists can officially declare a discrepancy. If theory and experiment do not match, physicists may have their mystery.

“Theory could be incomplete,” Bernard said. “We might be seeing new physics. Who knows?”

Theory through experiment

The least precisely understood ingredient of the theoretical prediction for g-2 is related to the strong force. Physicists cannot compute this number using quantum chromodynamics, the study of how quarks and gluons interact through the strong force. So they must find it through experiment.

Physicists from the BaBar collaboration at SLAC made this measurement by studying the production of a pair of subatomic particles called pions in collisions of electrons and their antiparticles, positrons, at a wide range of energies.

The physicists used a new technique to study collisions at different energies. They focused on electron-positron collisions in which the electron or positron released a high-energy photon, a particle of light, before colliding. The photons took different amounts of energy with them and therefore reduced the energy of the collisions by different amounts. So researchers were able to study different kinds of collisions without changing the energy at which they ran the accelerator.

The number of pion pairs created in these collisions depended on the energy of the collisions. Determining the relationship between the energy level and the number of events that resulted in pion pairs helped the physicists predict the value of g-2.

g-2, part II

Explanation of g-2 experiment / BNL

In the old g-2 experiment at Brookhaven, physicists fed spinning muons into a ring-shaped magnetic field almost 46 feet in circumference.

A muon’s spin axis changed by a set number of degrees at each turn as it circulated around the ring, like a horse on a merry-go-round that spins on its pole as the entire ride moves in a circle.

Eventually the muon would decay, breaking apart into an electron and neutrinos. The speed at which the escaping subatomic particles moved depended upon the direction in which the muon had been spinning.

Imagine you are about to fling yourself dramatically from a speeding car as it jumps the gap between two sides of an opening drawbridge. Do you try to jump backward to land on the near side of the bridge? Or do you try to jump forward onto the far side?

Even if you try to jump backward, the momentum you have from traveling in the speeding car will pull you forward. The force of flinging yourself backward will only slow your forward momentum, not reverse it, so you will probably land in the water below.

If you try to jump forward, however, your effort will combine with the forward momentum you already have, and you may just make it to the other side.

Similarly, a muon with a backward spin will not fling its electron as far forward as a muon with a forward spin.

Using this knowledge, scientists measured the energy of electrons to determine the direction in which their parent muons were spinning. Electrons with higher energy, the ones that made it to the other side of the metaphorical bridge, came from muons with forward spins. And electrons with lower energy, the ones that wound up in the water, came from muons with backward spins.

Some researchers have suggested physically moving the g-2 experiment from Brookhaven National Laboratory to Fermilab, in Batavia, Illinois, where facilities could produce a larger number of muons. Others have suggested designing a new g-2 experiment at KEK laboratory in Tsukuba, Japan. Both laboratories could be capable of finding more precise results.

Then, Bernard said, physicists could finally solve the mystery. Or create a new one.

Kathryn Grim

No Comments »

Manhattan Project opened door for future women scientists

April 17, 2009 | 7:50 am

Ellen Weaver made her first scientific discovery at the height of World War II. She made lead and aluminum “sandwiches” that blocked radiation from uranium fission, projects that led to the construction of the atomic bomb. For some women, such as Weaver, the war had opened up a window of opportunity in the sciences.

Speaking in February at the American Association for the Advancement of Sciences meeting in Chicago, Weaver recalled her own inspired start as a scientist in the midst of war, followed by her struggles for equality in the war’s aftermath. Addressing an audience including many women scientists and science teachers, Weaver praised the fact that women today are represented in unprecedented levels in science. She encouraged women to push for even further advancements in fields where they still struggle for equality.

“There’s no question that the situation is better for women today, but there still seems to be a dearth of women in science policy positions,” said Weaver, in a phone interview following the conference.

Back in 1944, Weaver complained about unequal pay for the handful of women working as chemists at the Manhattan Project facility in Oak Ridge, Tennessee. She noted that women scientists are still paid less than their male counterparts. According to the Association for Women in Science, women in the science, technology, engineering, and mathematical professions are paid less than men, even as their representation in those fields increased by 7 percent, to 26.1 percent, between 1983 and 2003.

Even though the Army denied Weaver’s request for equal pay, she excelled quickly in her job as an analytical chemist, by proving her capabilities in making radiation-blocking sandwiches.

That was Weaver’s first job out of college. Even though she had a few women colleagues, there was very little camaraderie among them, she said.

“We viewed each other with a bit of hauteur,” Weaver said.  ”I think it was a subconscious thing. I don’t think I had any desire except to be a woman in a man’s place,”

It wasn’t until after the war, when Weaver went on to graduate school in genetics and meet the only other woman graduate student in biochemistry at Berkeley, that Weaver had her “eureka moment.” She realized that women need to work with each other to end discrimination. After that point, she started to rely on the support of other women to maneuver her way in a man’s world.

“The Fifties were a terrible time for women,” she said.  ”My thesis advisor told me, ‘No woman had ever amounted to anything in genetics at Berkeley.”

Because of her, younger women did not face that same hurdle.

Weaver completed her PhD in genetics at Berkeley, and eventually became a genetics professor at San Jose State University.

She wrote the foreword to the book Their Day in the Sun: Women of the Manhattan Project (Temple University Press, 1999), which recounts women’s participation in the Manhattan Project.

By Kristine Crane

Symmetry Intern

2 Comments »

Can basic research bring renewable energies to scale?

March 12, 2009 | 9:13 am

The public is heaping pressure on the new presidential administration to increase renewable, domestic sources of energy. But the efforts of politicians and policy makers may amount to very little if science can’t catch up. Right now, our global energy future may rest in the hands of basic research scientists.

At the 2009 AAAS meeting, a session titled “Basic Research for Energy Security: A Call To Action,” featured speakers who drove home the idea that improving current alternative energy technologies—including silicon solar cells and lithium ion batteries—simply won’t be enough to solve the energy crisis. Instead, basic research scientists need to develop new technologies that are cheaper, more efficient, and fit for mass deployment.

“We’re going to need more than political will,” said Nathan Lewis of the California Institute of Technology, speaking about new technologies in solar panels. “We’re going need R&D and technology to get this to scale. Because the scale of energy is enormous, and therefore what works in your back yard or on your roof doesn’t apply to scaling globally, unless it’s very cheap and very amenable to mass deployment.”

Lewis estimates that to fully supply the US with solar energy by 2050, we would need to install solar panels on one million homes a day, every day, for the next forty years. Right now, the most ambitious plan of this kind is in California, where the government plans to put solar panels on one million homes over ten years. But not enough silicon is produced in the world each year to create that many solar cells. Many other solar cell materials would also fall short. To achieve even a portion of this goal requires variety and versatility.

Read the rest of this entry »

Calla Cofield

1 Comment »

Simulating cosmic evolution

March 3, 2009 | 8:12 pm

A still from a simulated animation of a Type 1a supernova. Click image to see a low-res version of the animation (avi format). Image and animation courtesy of flash.uchicago.edu.

What is matter? Where did it come from? What is the future of the universe? To answer these compelling questions, astrophysicists are trying to learn more about the physics of the big bang, and the origin of structure–the formation of the initial clumps of matter from the primordial soup. Computational tools and resources are indispensable to pursuing these fundamental questions.

Robert Rosner, director of Argonne National Laboratory and professor at the University of Chicago, spoke Friday, February 13 at the AAAS conference in Chicago about the role of simulation in studying the origins and evolution of the universe.

Direct observation of the cosmos has uncovered a host of facts. For example, the universe is expanding from the big bang and its expansion is accelerating. But observation will only take us so far, said Rosner. Scientists need to use theory to construct possible ‘scenarios’, and test them via experiments at particle accelerator laboratories and via computer simulations. Rosner presented simulated animations from a couple of important projects as examples of the use of simulation in astrophysics.

The Millennium Simulation Project is helping to clarify the physical processes underlying the buildup of real galaxies and black holes. It has traced the evolution of the matter distribution in a cubic region of the universe over 2 billion light-years on a side. According to its Web site, this simulation kept the principal supercomputer at the Max Planck Society’s Leibniz Supercomputing Center in Garching, Germany busy for more than a month.

The ASC/Alliances Center for Astrophysical Thermonuclear Flashes at the University of Chicago runs simulations to solve the problem of thermonuclear explosions on the surfaces of compact stars. Their simulations of Type Ia supernovae, exploding white dwarf stars, have shown that an internal flame ‘bubble’ emerges at a point on the stellar surface, leading to surface waves that converge at the opposite point, and causing a shock and subsequent detonation of the entire star. Previously, scientists thought that the original flame would directly transition to a detonation. Based only on well-known physical processes, these simulations exemplify the potential of numerical simulations for scientific discovery.

Robert Rosner. Image courtesy of Lloyd DeGrane.

Rosner noted that computing is moving towards the exascale (processing power of over 1018 FLOPS). He compared the current transition in computational methods and capabilities to the early 1990s when scientists moved from vector machines (such as the Cray) to massively parallel computers. At that time, the challenge was to modify existing codes optimized for vector machines to run efficiently on massively parallel machines; the new challenge is that of resource diversity across the network–how to construct algorithms that are flexible enough to efficiently exploit a variety of resource types, such as multi-core and heterogeneous processor computer architectures.

“For instance, we’re learning that for heterogeneous systems the MPI programming model, a standard for message passing between distinct tasks running concurrently on a computer, may not work well,” Rosner said. “Do we go back to multithreading and OpenMP from the early 90s? Are new algorithms needed? We may well have to change the way we program once again; and given our huge investments in existing codes, this will be a huge challenge.”

“The future is stunningly exciting,” said Rosner. “When we get to exascale computing we can capture the visible universe and we will understand how the observed structure came to be. We’ll be able to reach more reliable conclusions about the fate of our universe.”

by Anne Heavey

This story first appeared in International Science Grid This Week.

Guest author

1 Comment »