HPCwire: Carnegie Mellon Team Simulates Cosmic Evolution

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

Leading HPC
Solution Providers

HPCwire >> Topic >> Systems

July 06, 2007

Carnegie Mellon Team Simulates Cosmic Evolution

First-of-Its-Kind Model Incorporates Black Holes, Helps Predict Where To Aim Future Telescopes

PITTSBURGH, July 2 -- By incorporating the physics of black holes into a highly sophisticated model running on a powerful supercomputing system, an international team of scientists has produced an unprecedented simulation of cosmic evolution that verifies and deepens our understanding of relationships between black holes and the galaxies in which they reside. Called BHCosmo, the simulation shows that black holes are integral to the structure of the cosmos and may help guide users of future telescopes, showing them what to look for as they aim to locate the earliest cosmic events and untangle the history of the universe.

The research team is led by Carnegie Mellon University and includes scientists from the Harvard-Smithsonian Center for Astrophysics and the Max Planck Institute for Astrophysics in Germany. The research is in press with The Astrophysical Journal (http://lanl.arxiv.org/abs/0705.2269).

"Ours is the first simulation to incorporate black hole physics," said Tiziana Di Matteo, a theoretical cosmologist and associate professor of physics at the Mellon College of Science at Carnegie Mellon. "It is very computationally challenging, involving more calculations than any prior similar modeling of the cosmos, and the result offers us the best picture to date of how the cosmos formed."

Di Matteo performed her simulation using the Cray XT3 system at the Pittsburgh Supercomputing Center (PSC), the most powerful "tightly coupled" system available via the National Science Foundation's TeraGrid. Di Matteo's collaborators included Jörg Colberg at Carnegie Mellon, Volker Springel and Debora Sijacki at the Max Planck Institute for Astrophysics, and Lars Hernquist at Harvard.

Experimental observations reveal that black holes are important regulators of galaxy formation and, ultimately, the fabric of today's universe, according to Di Matteo. Nevertheless, previous simulations didn't take black holes into account because the computing demand was prohibitive.

"Including black holes in computer simulations is critical. The galaxies we see today look the way they do because of black hole physics," added Springel, junior research group leader at Max Planck. "We must do simulations to understand the role black holes have played in forming structures of both the early universe and today."

The largest black holes, called supermassive black holes, lie at the center of each galaxy. They can arise initially when the first stars collapse under their own gravity. Surrounded by dense gas in their central locations, they consume surrounding material, both gas and stars, and rapidly grow to become monstrous in size, some with masses a billion times that of our sun. But evidence suggests that supermassive black holes are self-regulators -- they don't feast forever and they never swallow a whole galaxy, Di Matteo said.

In her cosmic simulation, as in reality, galaxies routinely collide. The supermassive black holes embedded at the center of these galaxies choreograph the dynamics of galaxy collision. The result is a tremendous belch of energy produced as the merging black holes form a luminous state called a quasar.

"Quasar formation really captures when the fun happens in a galaxy," Di Matteo said. "You can only use a simulation to follow a complex, nonlinear history like this to understand how quasars and other cosmic structures come about."

Di Matteo's simulation covered multiple scales of time and space up to 100 million light years. It was impossible to do without a powerful supercomputer. "The XT3 is ideal for this simulation because it has incredibly fast built-in communication," she noted.

Di Matteo set up the simulation's initial conditions to reflect observed cosmic microwave background radiation produced at the birth of the universe. Then she seeded the simulation with a quarter of a billion particles that represented everyday measurable matter. For the simulation, Di Matteo used fluid spheres to represent chunks of matter such as cooling gas. This step was essential so the investigators could calculate all the physical forces on these chunks. She also factored in gravity exerted by dark matter, the unseen material that comprises 90 percent of the universe. Additionally, her calculations accounted for the forces associated with various cosmic phenomena, including cooling gas, growing black holes and exploding stars.

To make the computation possible, the scientists used 2,000 processors -- the whole system -- of the parallel Cray XT3 during four weeks of computing time. Even with this vast computing power, special techniques were needed just to be able to compute all the gravitational forces involved. For example, a "tree" was built-up in which nearby cosmic particles lay in the same "branch" and nearby branches were linked. By computing the forces on particles from this entire tree structure, the number of computations required was reduced by a factor of a few million to something manageable.

The results were impressive. Di Matteo's simulation allows a scientist to easily follow the collapse of galaxies. She has also resolved spatial scales that range from structures inside a single galaxy to the filamentary structures that the galaxies inhabit, which are tens of millions of light years long.

"We believe that our work has profound implications for cosmology," Di Matteo said. "We have found that the most massive black holes early on are not the most massive black holes we see today, so simulating the dynamic evolution of these structures is critical to understanding cosmic history. I want us to be bold enough to model the whole universe to scales observed with Sloan Digital Sky Survey (SDSS)." The SDSS is the largest light survey of the cosmos, which has catalogued nearly 100 million galaxies to date.

"With our simulations, we can predict what next-generation telescopes should see as they peer back 13 billion years to the time just after the Big Bang. We should be able to answer whether we are getting the universe right with our simulations and how it evolves as we go back in time," Di Matteo said.

Computing constraints currently hamper future work, but Di Matteo expects to run her next simulations on more powerful computers. She is also working with faculty from Carnegie Mellon's School of Computer Science to develop even faster ways of combining the physics of the very large with the very small in the same calculations using a set of tools called dynamic meshing.

For more information about Di Matteo's research, visit http://web.phys.cmu.edu/~tiziana/. For information on these simulations, including graphics, see www.psc.edu/science/2006/blackhole/ and http://web.phys.cmu.edu/~tiziana/BHCosmo. For information about PSC, see www.psc.edu.

The research was supported in part by the U.S. National Science Foundation.

-----

Source: Carnegie Mellon University

Article Tools

Share & Save Options

Discussion

There are 0 discussion items posted.

Join the Discussion

You must be a registered user in order to comment on HPCwire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCwire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Feature Articles

The New Science of Visual Analytics

When Jim Thomas set out to find new ways to deal with the mountains of information our society generates, he didn't just create a new organization, he created a new science. In this article we'll take a look at how the National Visualization and Analytics Center is transforming the problem of finding needles in haystacks into an opportunity for a more secure future.
Read More...

The Week in Review

ORNL Jaguar doubles its performance; the SC08 Cluster Challenge is gearing up; the University of Central Florida uses Army dollars to purchase an IBM super; and IBM's RoadRunner prepares to break the petaflop barrier. John West recaps those stories and more in our weekly wrap-up.
Read More...

Revaluating FPGAs for 64-bit Floating-Point Calculations

We now have generally available 2.5 GHz quad-core Opterons and Virtex-5 LX330, SX95T and recently announced SX240T FPGAs. In addition to this, Xilinx is releasing a new version of their floating-point cores that reduces the amount of logic and DSP slices needed for building floating-point function units. Taken together it is time to revisit Opteron floating-point performance versus FPGA performance.
Read More...

Read more HPCwire features...

Top Headlines

Parallel Processing Calls for a Fortress Mentality

May 14 | InfoWorld | Sun Microsystems is taking the lessons learned from Java and applying them to the application development challenges of the high performance computing realm. Read more...

IBM Set to Test the Fastest Computer in the World

May 14 | Computerworld | IBM is assembling the final pieces of what they hope will soon become the world's most powerful supercomputer. Read more...

IBM Shifts Cell to 65 Nanometers

May 13 | EETimes | IBM Corp. has announced the next-generation version of its Cell processor, the first specifically geared for computer servers. Read more...

Record-Setting Simulations on Ranger Reconstruct the Reionization Era

May 12 | Texas Advanced Computing Center | In the coming months, Dr. Michael L. Norman of UCSD will use Ranger, the world's most powerful supercomputer for open-science research, to perform the largest cosmological simulation to date. Read more...

2 AMD Executives Out in Restructuring Amid Slump

May 12 | BusinessWeek | Two executives have left AMD, including the head of the slumping chip maker's microprocessor division, as the company tries to engineer a dramatic turnaround to fend off larger rival Intel Corp. Read more...

Read more headlines...

Multimedia

Podcast: Interview with Ben Bennett of ClearSpeed Technology

Today, HPC organizations are requiring substantially more floating point performance to solve real-world problems. In this podcast, Ben Bennett, ClearSpeed General Manager, discusses how acceleration technology can improve the overall performance of standard x86-based systems...