In high-performance scientific and engineering computing circles—the playgrounds of weird science—petaflop supercomputing is coming. The pursuit of the petaflop is big. How big? How about ‘A New Arms Race”?

Willian R. Pulleyblank, the Director of Exporatory System Servers in IBM’s Research Division, and the Director of the IBM Deep Computing Institute, says…

“Computer simulation is being broadly recognized as a third pillar of research in science and engineering, joining Theory and Experimentation. However the resulting modeling requirements go far beyond the capabilities of current supercomputers. This takes us into the domain of terascale and petascale computing.”

According to this article from the New York Times, the world is involved in a new arms race though, this time, not in terms of armaments. Rather, the race is on to build the world’s fastest, petaflop computer.

A petaflop is 1,000 trillion floating-point operations per second. That’s 100,000 times faster than the average desktop computer, according to this article from the National Bioinformatics Institute. The Japanese government (as early as the fiscal year ending in March 2011)—as well as IBM in the U.S., through its BlueGene effort—are aiming for 1 quadrillion computations per second. That’s two million times faster than the average desktop PC.

“It’s becoming an issue of national pride,” said Steve Wallach, a supercomputer designer who is a vice president at Chiaro Networks, a technology provider for high-performance computing. “That’s where the Japanese are coming from, and now the Chinese want to be viewed as a Tier 1 country in every respect.” IBM’s Blue Gene/L—currently the world’s fastest computer, according to the NYT article—is installed at Lawrence Livermore National Laboratory.

According to this article at Top500.org, the U.S. leads with 294 of the fastest 500. Japan is on the list with 23 systems, while all other Asian countries combined add 58 systems (with 19 from China, up from 17 just six months ago.) Europe has 114 systems installed, and Germany was reclaimed the European top spot from the UK, with 40 systems compared to 32.

In the U.S., main applications for supercomputing are in the military. IBM’s Purple, for example, ‘… can perform sophisticated physics and engineering simulations to help researchers better gauge the safety and reliability of aging nuclear weapons in the absence of nuclear testing.’ (BUNMI: THIS NEEDS TO BE LINKED ONCE URL IS DETERMINED: Also see a companion article, Extreme Tech II: Building a Better Supercomputer Building) Such massive computing power can also be utilized in such areas as biotechnology, earthquake simulation, and others.

According to the NYT article, ‘Many scientists and technology executives in the United States are concerned about losing out in crucial markets like oil and gas exploration, automobile design and manufacturing unless they, too, have access to the fastest supercomputers.’

In April, ‘…the Japanese Automobile Manufacturers Association reported that the Earth Simulator had been used by automotive engineers to greatly increase the speed and resolution of car-crash simulations, potentially offering a significant reduction in development time for new car models.’ Some U.S. Executives have seen this with alarm, arguing that high-performance computing is now decisive in industrial competitiveness. Suzy Tichenor, vice president for high-performance computing projects at the Council on Competitiveness in Washington, says “This is a subtle but important change in the competitiveness game.”

Supercomputers are, in fact, used in real-world manufacturing applications. For instance, ‘Procter & Gamble used a supercomputer to study the airflow over its Pringles potato chips to help stop them from fluttering off the company’s assembly lines.’

I’m not sure that flying snack foods will require the computing power of the latest generation of petaflop computers, however, especially since—according to this article on Softpedia—they’ll cost $1B a pop. While potato chips are, indeed, a cosmic consideration of The Human Condition, that article mentions something of perhaps even greater importance. “These machines can be used to answer questions that literally will mean life or death of humanity,” said Peter A. Freeman, assistant director at National Science Foundation.”

Do you see a need for supercomputer-level horsepower in engineering applications? Should more such computing time be dedicated to industrial/manufacturing competitiveness? Could we be petaflopping ever closer to a Terminator-like, computers-taking-over-the-world scenario?