High-Powered Computing Center Boosts Big-Data Analysis

Published in Insight - June 2015

The Maryland Advanced Research Computing Center (MARCC) democratizes the big data revolution, giving researchers the computing muscle they need for enormous projects like analyzing human genetic activity, creating three-dimensional maps of the universe and peering into the human heart.

Natalia Trayanova and her computational cardiology team are using MRI scans, heart-specific proteins and other data to create highly detailed, digital simulations of patients’ hearts. Zooming in to the molecular level or out to see the entire beating organ, researchers can pinpoint problem areas and test different treatments on a computer screen — considerably less invasive than an operating table.

This complex work, which could one day transform cardiac care, takes place at computing centers on the Homewood campus of The Johns Hopkins University. Some of the simulations run on many processors at once, so the work often has to wait, sometimes for weeks, until a sufficient number of machines are available at the same time, says Trayanova, who holds joint appointments in the school of medicine and Whiting School of Engineering.

That’s why she can barely contain her enthusiasm for last month’s opening of MARCC (pronounced “MAR-see”) at Johns Hopkins Bayview Medical Center. “Excitement is too mild a word,” she says.

Built with $30 million from the state of Maryland, MARCC is the largest high-performance computing center in a mid-Atlantic academic setting, with more than 19,000 processors and 17 petabytes of storage capacity — equal to more than 20,000 home computers.

The 3,786-square-foot center is shared by the Johns Hopkins University School of Medicine, Bloomberg School of Public Health, Krieger School of Arts and Sciences, Whiting School of Engineering, and the University of Maryland, College Park. Fiber-optic cables connect the campuses at a speed of 100 gigabits per second, up to 10,000 times faster than a home Internet connection. Eighty percent of MARCC’s computing power is already spoken for, but the Johns Hopkins Bayview lot has space for four additional modules.

The goal is to put high-powered computing within reach of all researchers. For Trayanova, that means her cardiac simulations can proceed without skipping a beat.