Dr. Fernando Cucchietti

A Supercomputer From Networked Macs

Using Mathematica, Dr. Cucchietti created this image of a quantum state of the Kuramoto Model, displayed in the representation known as the Wigner function. For ease of integration into Keynote presentations and other reports, he copies Mathematica files into Preview and saves them as PDF files.

In 2005, Dr. Cucchietti’s group got funding for a departmental supercomputer, but once they estimated the refrigeration and extra power and additional office space they’d need, there was no money left for the computer.

At about this time, Dr. Cucchietti read about Apple Xgrid technology built into Mac OS X and how it could turn a group of networked Macs into a low-cost supercomputer. In addition, he learned it was easy to set up and administer and didn’t require any special modifications to the lab. So his department used the money to upgrade the computers in the group to top-of-the-line Macs that Dr. Cucchietti used, remotely, to run his simulations in the background using Xgrid. And, in 2007, he gained even more supercomputing power for his work when they upgraded their Macs to Mac Pro computers.

“I was really proud of having solved all the problems at once, having gotten a big computer for my computations, and having spent almost no time configuring everything.”

Using Xgrid, Dr. Cucchietti can use the collective power of his colleagues’ Macs when he needs to, yet it doesn’t prevent his fellow scientists from working independently on their own projects. “With the kind of computing I do, I can use this extra technology very efficiently,” he says. “That’s because I don’t do a lot of communication-intensive parallel computing. My processors can talk to each other sometimes maybe once or twice a day.”

“I was really proud to have solved all the problems at once, having gotten a big computer for my computations, and having spent almost no time configuring everything,” he adds. “Xgrid was great because I didn’t have to do anything. I set it up and it was working from the first day. It was very easy.”

“I was solving the problem of space because we didn’t have to get a new office for the computer, I was solving the problem of power because it was already in each office. There was no problem with refrigeration, and everyone was getting a really fast computer. And because the simulations weren’t going to be running all the time, they could take advantage of that too—people could submit their own jobs and run them.”

Finding Solutions Without Quantum Computers

Will greater computer efficiency in the lab today help speed the creation of the quantum computer of the future? Dr. Cucchietti believes it will likely be several decades before fully realized quantum computers could be ready for use—if they’re even possible at all. “Some people say it will never happen," he says. "I am one of the optimists.”

But even if quantum computers remain elusive, says Dr. Cucchietti, physicists are still making important and useful discoveries in pursuit of this enigmatic goal. In fact, his primary area of research today may help pave the way for scientists to solve more complex problems in the near future rather than wait decades to do so. Using Macs and Xgrid, he’s designing laboratory experiments in which a quantum computer won’t even be necessary to simulate quantum problems. "These experiments can be done today, so people won't have to wait ten or twenty years," he says. "It’s a very big thing for physics because it will open up the ability to solve a lot of old problems we haven't been able to because we didn't have the incredibly huge computers to solve them. Now we'll be able to solve them right in the lab."

< Previous Page 1 2