NASA, Intel and SGI Plan Linux Supercomputer
Silicon Graphics (SGI) announced that NASA has chosen the company's Altix as the foundation of Project Columbia, a collaboration between SGI and Intel that is expected to foster scientific breakthroughs in space exploration, global warming research and aerospace engineering.
For the project, NASA plans to integrate a total of twenty 512-processor Altix systems with a 500-terabyte storage system to create the Space Exploration Simulator. The system will be among the world's largest Linux OS-based supercomputers.
"For over 20 years NASA and SGI have worked very closely and very successfully to deliver a series of technological firsts," said NASA Ames Director G. Scott Hubbard in a statement. "Project Columbia will build on the knowledge we gained from this close relationship to underpin the most ambitious missions in NASA's history."
Penguin in Space
NASA has noted that the use of Altix will boost its computing capacity ten-fold, and that the NASA Advanced Supercomputing Facility will be able to handle such critical projects as simulating future space missions, projecting the impact of human activity on weather patterns, and designing safe and efficient space exploration vehicles and aircraft.
The collaboration between the agency and SGI builds on an eight-year partnership that has seen the development of the world's first 512-processor Linux server, which was based on standard microprocessors and open-source technology. It was named "Kalpana" after Columbia astronaut and NASA Ames Research Center alumna Kalpana Chawla.
SGI has noted that it has already delivered the first three installations to NASA, and the complete array should be finished within months. It also added that the Altix has set many performance records in the past, demonstrating ability to run manufacturing, engineering and scientific applications in a Linux operating environment.
John Parks, Deputy Chief of the NASA Advanced Supercomputing Division noted in a statement, "[W]e expect the Space Exploration Simulator to initiate a new era in high-performance computing, one in which operating a 10,000 processor supercomputer may be as straightforward as managing twenty workstations."
Growth of Linux Supercomputing
NASA is not the only entity to embrace Linux for supercomputing needs. In the May edition of the TOP 500 Fastest List, which measures the speed and dominance of the world's supercomputers, Linux was represented on over half the list.
This is in contrast to past years, said list cocompiler Erich Strohmaier in an interview with LinuxInsider. He pointed out that five years ago, only a few Linux machines appeared on the list, but that its presence has been growing strongly.
Strohmaier credits the rise in Linux-based supercomputing to a number of factors, including lower cost, its effectiveness in cluster systems and the establishment of more performance benchmarks.
He added that putting Linux together with Intel processors, as SGI and NASA is doing, has also proved to be a trend. "We see quite a strong rise in Linux-based cluster systems with Intel or Opteron processors," he said. "This is happening especially strongly in the academic community, which always has a need for faster processing power."
The rise of Linux supercomputing appears in some commercial and industrial areas, but it is primarily in the scientific community that its presence is being established. Like the NASA project, other scientific entities and academic labs are investigating what Linux can do.
For example, the Department of Energy's Oak Ridge National Laboratory (ORNL) announced a partnership with IBM to draw on its Linux-based Blue Gene technology in order to build the world's most powerful supercomputer by 2007.
Thomas Zacharia, ORNL associate laboratory director for Computing and Computational Sciences, told LinuxInsider that the lab appreciates the flexibility of Linux.
"This will allow us to explore a range of applications and technologies," he said. "Having more supercomputing power will give all kinds of researchers greater ability to change how science is done."