Frequently Asked Questions
HPC in the United States3 questions
[ + ] Why use High-Performance Computing?
One benefit is that high-performance computing speeds up the process of innovation. To quote Deborah Wince-Smith of the U.S. Council on Competitiveness, supercomputers "offer an extraordinary opportunity for the United States to design products and services faster, minimize the time to create and test prototypes, streamline production processes, lower the cost of innovation, and develop high-value innovations that would otherwise be impossible." Furthermore, “Supercomputing is part of the corporate arsenal to beat rivals by staying one step ahead of the innovation curve. It allows companies to design products and analyze data in ways once unimaginable.” To learn more, read the Council on Competitiveness' piece "The New Secret Weapon." Also, visit our How HPC Can Help blog post.
[ + ] How does the United States compare to other countries in terms of high-performance computing capabilities?
From the TOP500 list, which uses the LINPAK benchmark, the United States has five computing systems in the top 10 and has 263 systems in the top 500 computing systems in the world.
[ + ] Who uses high-performance computing in the United States?
Users of high-performance computing in the United States include scientists at U.S. Department of Energy national laboratories, companies and academics. For instance, companies such as Siemens, Ford and Goodyear have benefited greatly from high-performance computing. To learn more, visit our Success Stories page.
Performance and Capacity4 questions
[ + ] What are the basics of High-Performance Computing?
High-performance computing (HPC) is a term used to describe the computation of an extremely large number of calculations in a practical amount of time. The two elements required to do this are a high-performance computer (with a speed in the teraflops range or greater) and software code that instructs the computer how to accomplish the calculations. A traditional desktop computer has a single Central Processing Unit (CPU) and calculations are completed by breaking down a problem into a discrete series of instructions. The instructions are executed by the computer one after another where only one instruction can be followed at any moment in time. The hardware in High Performance Computers consists of multiple CPUs and is configured to run parallel computations. In parallel computing, software directs the solution of many calculations at the same time for one large problem. Each problem must be broken down into discrete parts that can be solved concurrently with the instructions written for each piece to be solved simultaneously on different CPUs. Learn more.
[ + ] What is a petaflop? What is a teraflop?
A petaflop is 1015, or 1,000,000,000,000,000, floating-point operations per second. A teraflop is 1012, or 1,000,000,000,000, floating-point operations per second.
[ + ] How fast are these supercomputers?
The fastest supercomputer in the world by the LINPAK benchmark is currently the K computer at the RIKEN Advanced Institute for Computational Science in Kobe, Japan (source). To achieve this status, the K computer operated at 10.51 petaflops in November 2011.
[ + ] What's the difference between high-performance computing and other forms of computing?
High-performance computing ties together the strength of a large number of processors to solve highly calculation-intensive tasks. Personal computers and clusters of computers can solve small scales of these same problems, or sections of these problems; however, it is the hardware (physical pieces of the computer) and the software (the instructions) that differ and that allow high-performance computing systems to solve the massively large and complex tasks quicker and with more details.
Working with LLNL3 questions
[ + ] How much does it cost to run or use a high-performance computing system?
The cost depends on the time and expertise needed to develop the software program; the availability of a computer; the amount of power needed to run the computer; and the cost of the computer.
[ + ] Who runs these computers?
The computers are maintained and operated by a team of technicians and computer scientists on location. Models and simulations run on these computers are developed by teams of computer scientists, engineers and domain scientists for any given project. Access to computers is available locally or remotely.
[ + ] How do I contact a national laboratory if my company is interested in harnessing its high-performance computing capabilities?
Many national laboratories are partnering with industry. One way to access the national laboratories is through the High Performance Computing Innovation Center. To learn more about ongoing projects where the national laboratories are partnering with industry, visit the HPC for Energy Incubator page.