Purdue University, located in West Lafayette, Indiana, has some talented technicians who decided they wanted to assemble their own supercomputer in a very short time frame. Their expectations were high: they planned to build it in one day.

On Monday, May 5th, the Purdue technical team, along with some help from other technicians from nearby rival Indiana University, started unpacking boxes of 812 Dell servers at 6 A.M.  By 1 P.M. that same day the teams had 500 of the 812 supercomputer nodes operational and were running 1,400 jobs for campus researchers, accordingly to Purdue’s Vice President for Information Technology and CIO, Gerry McCartney.

"The assembly was finished much faster than we expected, and by noon we were doing science," said McCartney. "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…"

"We discovered that a build like this leverages the commodity nature of cluster computing by using standard computing parts. 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 supercomputer has an appropriate name: it is named “Steele” after Purdue University Computer Center’s retired director, John Steele. Purdue said that Steele’s performance will put it in the top 40 rank of the world’s Top 500 supercomputers. The supercomputer has been designed to perform 60 trillion operations per second.

Steele is the largest supercomputer on a Big Ten college campus that is not part of a national research center; it was funded by Purdue faculty members who contributed research funds rather than procure equipment for their own laboratories.

Simply phenomenal, is all we can say.