When professor Dave Kaeli approached them about entering a supercomputer competition, electrical and computer engineering students Neel Shah and Tushar Swamy had zero experience in the task at hand. “I had never even built a regular computer,” said Shah, let alone a super one. But that didn’t stop them from signing up…nor from stealing the show entirely.

In collaboration with three students from Bentley University, the Northeastern duo built a supercomputer within the necessary financial and power constraints laid out by the Student Cluster Competition. And then, over a 48-hour period last week, they let their computer run five data-intense applications in a neck-and-neck race to the win.

Seven other teams, hailing from as far away as Australia, also competed in the challenge. But one thing set Northesatern’s approach above all the rest. “We chose the most unconventional hardware out of all the other teams,” said Shah. Instead of building their computer with central processing units alone, they also used graphical processing units.

Both Shah and Swamy had four years of undergraduate research experience in Kaeli’s lab where they learned the ins and outs of the GPU. This specialized electronic circuit was designed for processing images, as its name implies, but in Kaeli’s lab they’re also using it to process massive data sets in parallel. The approach didn’t only give them good power efficiency, it also brought their costs down.

“There used to be something called the Top 500 supercomputers,” said Swamy. “But they were extremely power hungry. So now there’s the Green 500, too.” In keeping with that overarching shift in the community toward more efficient high performance computing, this year’s competition had two tracks for the first time ever: one restricted only by power, and one restricted by power and cost. Shah and Swamy’s team entered the latter, or “commodity” track.

To achieve high performance without spending a lot, they used a low-cost advanced platform called an APU, or acceleration processing unit. Made by Advanced Micro Devices, the APU combines both CPUs and GPUs on a single board. The approach won them first place in the commodity class  but it also stood up rather well against the standard teams: When competing against computers that cost nearly a million dollars, they said, the Northeastern/Bentley computer came in fourth place.

While the actual competition was a harrowing 48 hours of sleeplessness, spiked with moments of hair-pulling stress, like when they discovered in the middle of one run that it wouldn’t finish in time, the whole endeavor involved a much larger commitment. The team spent months working on their computer, communicating with the Waltham-based Bentley students using things like Google hangouts along the way.

Before they arrived at the event in Denver, they’d made sure the computer was able to successfully process all of the applications they knew would be in the competition, as well as readying it for a mystery program that wouldn’t be revealed until they set foot in the conference hall.

They also made sure to keep a little bit of the computer’s processing power available during every run to use for queueing up the next run. Something in this combination of good time-management skills, cross-institutional collaboration, and thinking outside the hardware box bought them their success, said Swamy.

And while they may not have had previous experience building computers from scratch, Shah and Swamy don’t plan to stop now. They found out during the competition that they’re proposal for the international supercomputing cluster challenge in June was accepted. Next stop, Germany.