A number of new and powerful machines are being built and installed for federal agencies.
Supercomputers are built of heaps of processors and can perform billions and trillions of computations per second, with power far greater than what traditional computers offer.
They enable the processing of massive amounts of data, new breakthroughs in artificial intelligence, and future-facing modeling and simulation research that help the nation’s top experts build analyses and formulate predictions across many fields. For those reasons and others, the United States has for years been steadily investing in these machines of the future—and this year that continued.
Here’s a rundown of some of the new high-performance supercomputers the government bought and developed in 2021, and the research they’ll underpin.
Driving New Discoveries About Earth
The advanced systems offer many applications, including helping scientists to advance fundamental research into the climate and the interconnected nature of the Earth as a complex system.
In January 2021, the National Center for Atmospheric Research announced plans to unleash a powerful new supercomputer to study phenomena like solar storms, climate change, and extreme weather events that threaten humans’ home planet. Built and installed by Hewlett Packard Enterprise at a government facility in Wyoming, the machine was said at the time to cost at least $35 million.
The supercomputer will be used to investigate concepts in the emerging arena of Earth system science, through which scientists explore the planet as a whole system and hone in on its interconnected physical, chemical, and biological elements and processes that can drastically impact humanity. Among a variety of topics, the new system is expected to allow for more accurate models for the simulation of the subsurface flows of water, oil, and gas; expanded computing power for better representations of physical processes to help forecast wildfire threats and behavior; and deep, three-dimensional simulations of the Sun's plasma flows and magnetic fields supporting solar storm predictions for weather events that can disrupt communications means and power grids.
Months after that announcement, in the summer, Lawrence Berkeley National Laboratory unveiled Perlmutter: a next-generation supercomputer that’s poised to run among the most powerful globally. At that time, officials noted that the system will support the assembly of “the largest 3D map of the visible universe to date.”
And more recently the Energy Department’s National Renewable Energy Laboratory revealed it is set to gain a supercomputer—Kestrel—that will support huge, data-intensive workloads and necessary research and development to help the nation prepare for future clean energy-aligned needs.
The lab’s computing applications span common themes including materials discovery, process modeling, fluid dynamics, resource mapping, and the analysis of large-scale systems with real-time optimization. Kestrel will eventually allow for new and advanced modeling, simulation, AI and analytics capabilities around those areas, and it’ll boost research across the lab’s portfolio.
Support for the Nuclear Stockpile
To safeguard and modernize the U.S. nuclear stockpile, the government primarily uses science-based assessments that do not involve nuclear explosives testing, which the nation voluntarily ceased in the 1990s.
Supercomputers help support that crucial work.
In April, Los Alamos National Laboratory announced the procurement of a new system for researchers to use to embark on new discoveries across climate, diseases, materials—and particularly, nuclear deterrence.
At the time, officials said it was anticipated to be a first-class resource for LANL’s artificial intelligence, machine learning, simulation and other workloads, and an asset through which scientists in the lab’s weapons program can try new concepts out that are associated with its ongoing and emerging endeavors.
Later this year, Dell technologies also announced that it was selected to deliver a notable amount of expanded computing capacity at Lawrence Livermore, Los Alamos and Sandia National Laboratories.
That award was worth at least $40 million and officials confirmed during the announcement that the fresh upgrade in capabilities is expected to enhance research and development activities to maintain the safety, security and effectiveness of the nation’s nuclear weapons stockpile.
The Defense Department also uses such sophisticated machines to meet its missions, and runs its own High Performance Computing Modernization Program.
This year, DOD confirmed that it awarded California-based Penguin Computing two contracts worth a combined $68 million to provide two high-performance supercomputers and associated capabilities to the Navy and Air Force.
The systems and software are expected to “significantly enhance the [department’s] ability to tackle the most demanding and computationally challenging problems in fluid dynamics, chemistry and materials science, electromagnetics and acoustics, climate/weather/ocean modeling and simulation, among other applications,” officials said in their announcement.
Once fully installed, those advanced capabilities will be accessible to all services and DOD subcomponent agencies.
Telecommunications company Verizon was also awarded a DOD supercomputing-related contract in 2021. It’s worth up to $495 million over ten years for secure, high performance wide area network services in support of a supercomputing network that links together 200 science and engineering labs and high performance computing locations.