n recent months, scientists
at the Pittsburgh Supercomputing Center reported progress in the fields
of genomics, public health, chemistry and machine learning thanks, in
part, to a new National Science Foundation-funded supercomputer –
Bridges. The system, which was also used to model the possible benefits
of influenza vaccine choice, provided compelling results that suggest
adopting a vaccine choice policy could curtail the
national cost and enormous annual burden of influenza in the United States.
Federal
funding and support for supercomputers or high-performance computing
(HPC) projects, such as Bridges, is critical for enabling new scientific
discoveries and cures for diseases, enhancing national security and
cyber defenses and improving economic competitiveness. Yet, U.S.
leadership in this market is being contested as other countries are
beginning to invest aggressively as they recognize the technology’s
strategic importance.
Think
of it like the Space Race of the 1960s, but with investment going into
software and large-scale HPC systems rather than rockets. The stakes are
different, but the potential for world-changing results is just as
important.While the U.S. continues to hold a prominent position in HPC, the
Associated Press recently reported that
China has now officially displaced the U.S. for the first time with the
most supercomputers on the Top 500 listing of the fastest
supercomputers in the world. Fifteen years ago China didn’t even have
one HPC on the Top 500 listing.
According to the International Data Corporation, the
Asia Pacific region had a HPC growth of 15 percent in 2015, where the
U.S. was only at 5 percent. The European Union has recently doubled
their investment in supercomputing research, and Japan’s High
Performance Computing Initiative program shows they are focused on
building an exascale supercomputer by 2020,
whereas the U.S. is currently on target for 2023.
The
reality is that HPC systems are no longer a technology reserved for
elite academic institutions, but are now bleeding into the enterprise,
disrupting industries and impacting our nation’s competitiveness and
innovation. For example, imagine a wind turbine. Without HPC,
researchers could only test for very general system-wide improvements.
With it, researchers can use HPC systems to handle massive amounts of
data and complex calculations to analyze the impact of friction at
multiple points on a single turbine blade, making improvements a
millimeter at a time. Efficiency gains can multiply quickly at that
level, leading to a whole new paradigm for investing in clean wind
energy.
Organizations like the Pittsburgh Supercomputing Center
are also pushing the HPC envelope with new and innovative architectures.
Bridges represents a new approach to supercomputing that focuses on
research problems that are limited by data movement. In addition to
serving traditional supercomputing users, Bridges will help researchers
tackle new kinds of problems in genetics, the natural sciences and
social sciences, where scientists are impacted by the volume of data
rather than computational speed. This project highlights another
innovative way HPC can help solve pressing issues facing the U.S.
But
we can’t stop there. Given the explosion of vast amounts of data and
the increasing importance of simulations in scientific research, we need
to make the next exponential leap in HPC, exascale computing and next
generation computing. To put in perspective, an exascale supercomputer
could operate faster than
50 million laptops. As
with the semiconductor industry, the country or region that leads HPC
and exascale will capture the related economic and societal benefits.
The U.S. simply cannot afford to fall behind in this race.
While
the private sector continues to make significant HPC-related
investments, the federal government also plays a vital role in advancing
HPC research. HPE recently participated in an Information Technology
Industry Foundation (ITIF) panel on Capitol Hill to urge Congress to
fully fund the Administration’s FY2017 request on the National Strategic
Computing Initiative (NSCI). The NCSI names three lead agencies,
Department of Energy (DOE), Department of Defense, and National Science
Foundation to lead these efforts.
Specifically, DOE plans to
establish, over the next 15 years, a viable path forward for future HPC
systems, and to research new progressive technologies that perform
“beyond Moore’s law,” including the all-important exascale computing.
We
continue to urge Congress to fund these critical investments, so the
U.S. can enjoy the greatest rewards of HPC-driven technology and give
our nation the advantage when attempting to solve the world’s most
pressing challenges.
http://thehill.com/blogs/congress-blog/technology/288315-strategic-investments-in-high-performance-computing-enhance-us
