Why America Must Win the Supercomputing Race
By Eric Isaacs
Courtesy: U.S. Department of Energy
Four decades ago, NASA put a man on the moon using a computer system less powerful than the electronics in many modern-day toasters. With that audacious act of technological faith, the United States took a giant step toward global leadership in science, engineering, and a myriad of other sectors that had not yet been imagined.
This week, when a Chinese machine was ranked number one on the most recent Top 500 list of the world’s mostpowerful supercomputers, the United States has lost more than international bragging rights. By creating the Tianhe-1A, with 1.4 times more muscle than America’s fastest supercomputer (the Jaguar at the Department of Energy’s Oak Ridge National Laboratory,) the Chinese have sent a forceful message to the world about their ambitious vision of their country’s scientific, economic and military future.
The United States cannot afford to take a back seat in computer technology to the Chinese, or to anyone else. The nation that leads the world in high-performance computing will have an enormous competitive advantage in every sector, including national defense, medicine, energy, environment, finance, manufacturing and product development.
But more important, the nation with the best and fastest supercomputers will attract the best scientific and engineering talent from around the world. And if the United States loses a generation of our top technological talent to another nation, we will feel the impact of that loss for decades to come.
There is no reason for the United States to yield its position as the world’s leader in supercomputing. As the director of Argonne National Laboratory, home to one of the world’s fastest supercomputers, I know that we already have in place a roadmap to create the next generation of supercomputers – systems that will be 1,000 times more powerful than the new Chinese machine.
Unlike today’s supercomputers, with speeds that are measured in petaflops -a quadrillion sustained floating-point operations per second – the next generation will be measured in exaflops – a quintillion, or one million trillion floating point operations per second.
These exascale supercomputers, as they are known, will be powerful enough to simulate worldwide climate change – or the extraordinarily complex functions that take place within a single human cell. Hundreds of American scientists and engineers at universities, in private industry and at our national laboratories already are racing to design the elements of an exascale system, re-imagining hardware, programs, and applications for a supercomputer that will come on line in 2020.
But our scientists cannot meet that deadline without a substantial national investment – in the future of leadership computing, and in American science overall.
We need to make sure that American researchers and engineers have access to the supercomputers and other technological tools they need to help solve the great scientific, energy, environment, and security challenges of our time. We also need to make sure that our laboratories are equipped with cutting-edge facilities that will draw talented young scientists from around the world.
We have no time to lose. Ten short years ago, China had no high performance computing ability. Today, after investing billions of dollars in its computer technology, China has 24 computer systems on the list of the world’s 500 most powerful supercomputers.
As we watch China take over the lead in supercomputing speed, Americans can take some comfort in the knowledge that the Chinese system is built largely from American-designed components. But the networking technology that brings those components to life was designed in China – and the Chinese already are at work on a 1-petaflop supercomputer made from Chinese parts.
America needs a substantial, longterm national investment to speed our journey down the road to exascale computing – a road that leads to economic growth, international competitiveness and national security. Without that commitment, the American supercomputers of the future may be labeled, “Made in China.”
Eric D. Isaacs is the Director of Argonne National Laboratory.