_This week, PopSci is peeking under the hood of some of the nation’s biggest and baddest supercomputers–the machines that turn big data into big discoveries, big technologies, and big leaps forward. Over the last week, we managed to get each of the busy machines in this series on the phone to see what they were up to on during a particular day. They were happy to share.
Today we chat with the 63 thousand processing cores of the University of Texas’s Ranger.
World Ranking: 17
Vital Stats: Performance: 579.4 teraflops. Inside: 3,936 quad-socket, quad-core Sun Constellation blades with 62,976 2.3 GHz AMD Opteron Barcelona cores and 123 terabytes of aggregate memory. There’s 1.7 petabytes of total disk space.
So What Are You Working On?
- A half-hour job to test the scalability of an application designed to model the Earth’s atmosphere. This is the kind of job-before-the-job that supercomputers do all the time. When researchers are doing big, boundary-testing science, there’s often no precedent from which to work. To test the viability of this particular atmospheric model, they have to get it on the system and see how it functions. The resulting tweaks and modifications made to the software ensure that researchers can scale it to even more cores than the roughly 33,000 that were required today. More cores means bigger science, with bigger potential payoffs.
- Modeling turbulent flow of the magnetized gases in the interstellar medium. It’s one in a series of simulations aimed at building a bigger picture of how stars form and why galaxies move the way they do.
- And spending an hour modeling radiation dosing for patients suffering from cancers. The way a radiation dose is delivered–its size, shape, duration, location, etc.–has everything to do with its efficacy. Moreover, this kind of modeling can help to reduce the side effects of radiation exposure to healthy tissues.
That’s not the only biomedical science Ranger will perform today. As part of a two-day job, Ranger models a protein linked to Alzheimer’s to help out researchers in pursuit of a treatment. Much of what proteins do is tied up in their shapes and their motions. By helping them model those things today, Ranger is helping define targets for potential Alzheimer’s therapies.
Catch up with more supercomputers here.