Supercomputer Gleans Knowledge from Massive Data Sets
Is this new cardiac medication helpful or lethal? How can we design head gear that effectively reduces the number of soldiers, athletes, and everyday citizens who suffer concussions? Will this new vaccine head off an outbreak in those who are inoculated?
In the past, those fundamental questions have been answered only through trial and error, enlisting willing human subjects to test potential solutions. Mountains of data are collected then painstakingly analyzed for clues. But, with the advent of high-performance supercomputers, scientists at the University of Rochester can begin to extract knowledge quickly and reliably out of large, dense data sets, by creating computer models that predict the effectiveness of various interventions or allow the scientist to test various solutions through precise computer simulations. Thanks to a partnership with IBM, the University of Rochester is at the forefront in bringing powerful computer capability into medical labs.
Earlier this month, New York State Governor Andrew Cuomo and officials from the University of Rochester and IBM opened the doors on a new UR facility that will bring new computing wizardry to solve health care’s most complex problems. The Health Sciences Center for Computational Innovation (HSCCI) is home to IBM’s next-generation supercomputer – the Blue Gene/Q – making it one of the five most powerful university-based supercomputing sites in the nation.
“With this partnership, New York has an unprecedented opportunity to participate in cutting-edge healthcare research while creating hundreds of jobs at the University and in the Finger Lakes region,” said Governor Cuomo. “This is the type of partnership we are starting to see throughout the state and I look forward to working with the private sector to transform New York into the best place to start and grow businesses.”
“Until now, we haven’t had tools powerful enough to extract the knowledge out of very large, very dense data sets generated by some of our experimental clinical studies,” said David Tompham, Ph.D., HSCCI director. “The concept of taking high performance computing technology and applying that to health research questions is what enables us to make sense, to create knowledge out of that data.”
Renovations to the University’s data center and acquisition of the Blue Gene/Q were made possible though a $5 million award from New York State at the recommendation of the Finger Lakes Regional Economic Development Council. Last year, the council identified the HSCCI as a priority project. The Center for Governmental Research estimates that the project could create 900 jobs and in the community and generate $205 million in new research funding over the next 10 years.
“We are deeply grateful to Governor Cuomo, Lt. Governor Duffy, our delegation in Albany, and the community leaders who have supported this transformative project,” said Joel Seligman, president of the University of Rochester.
The Blue Gene/Q is the pinnacle of high performance computing and is 15 times more powerful than IBM’s previous generation supercomputer. It is also the world’s most energy efficient computer system. Scientists at the University of Rochester are already applying high performance computing to research programs in vaccine development, brain injury, and cardiac disease.