Quantcast
Jaime Combariza, a Johns Hopkins computational chemist, is director of the Maryland Advanced Research Computing Center. (Will Kirk / Johns Hopkins University)
Jaime Combariza, a Johns Hopkins computational chemist, is director of the Maryland Advanced Research Computing Center. (Will Kirk / Johns Hopkins University)

JHU and UMD crave more (processing) power

Seventeen million gigabytes.

That’s the storage capacity of a new, high-capacity, 19,000-processor computing center set to open later this month at the edge of the Johns Hopkins Bayview Medical Campus.

The facility, known as the Maryland Advance Research Computing Center, or MARCC, or “mar-see,” will offer its immense processing power to researchers from the Johns Hopkins University and the University of Maryland, College Park.

But thanks to fiber-optic cable connections, those researchers will never have to leave their campuses, according to a statement from JHU.

While in the past, individual research groups at the schools might have had to use their own time, money and space to computing center, participants will now share the cost of running the center, according to the university.

So how for does 17 million gigabytes (that’s 17 petabytes for the computing cognoscenti) go? According to JHU, one genetic experiment comparing two types of tissue can generate up to 40 gigabytes of data, a simulation of heart function can generate one terabyte (or 1,000 gigabytes) of data, and an MRI or CT scan can generate up to two terabytes.

MARCC, supported by $30 million in state funds, is designed to make studies of that sort of data move faster, and 80 percent of the computing power is already allocated to research projects. If there’s enough demand — and funding — there’s room on the campus to build four additional computing centers, according to the university.

Leave a Reply

Your email address will not be published. Required fields are marked *

*