Technicians at Purdue University wanted to assemble their own supercomputer, and they had some high expectations. Apparently, they could have set them even higher.
The technical team set out yesterday to build a supercomputer made up of 812 Dell servers in a single day. Instead, they had it up and running by lunchtime.
"The assembly was finished much faster than we expected, and by noon we were doing science," said Gerry McCartney, Purdue's vice president for information technology and CIO, in a written statement. "The staff was enthusiastic, the weather was great, and there were no problems installing the hardware or software. There is no cloud to accompany this silver lining."
The Purdue team, with the help of a crew of technicians from athletic rival Indiana University, began unpacking boxes of servers at 6 a.m. Monday. By 1 p.m., they had 500 of the supercomputer's 812 nodes operational, and were already running 1,400 jobs for researchers across campus, according to a statement from the university.
Named "Steele" after John Steele, who retired as the Purdue University Computing Center's director, the supercomputer is designed to perform 60 trillion operations per second.
"We discovered that a build like this leverages the commodity nature of cluster computing by using standard computing parts," McCartney said in the statement. "By using commodity computer servers to build our supercomputer, we didn't have to fly in engineers or hire specialized technicians. We were able to do it with our own IT staff in about four hours."
The university said that Steele is the largest supercomputer on a Big Ten campus that isn't part of a national research center. The school also noted that its performance would put it in the top 40 of the current ranking of the world's Top 500 supercomputers. The Top 500 ranking, which comes out twice a year, is due to issue a new listing June 17 during the International Supercomputing Conference in Dresden, Germany.
The supercomputer was funded by Purdue faculty members who contributed research funds instead of purchasing equipment for their own laboratories, according to the university.