Cover Photo: The Cray 2 Supercomputer | NASA (Public Domain)
As a young boy, I would wake up early on Saturday morning and ignore the cartoon-laden television. Instead, I would run down a flight of stairs to my parents’ basement and climb a bar stool to reach the object of my desire: A Radio Shack TRS-80 personal computer. With my feet dangling over the side of the seat, I would flash up the silver and black monstrosity to play a game called ‘Zork’ – one of the few games available at the time.
Created by a couple of guys at the Massachusetts Institute of Technology (MIT) in the late 1970s, Zork was a text-based, choose-your-own-adventure that took the player through a world of dungeons and dragons. For a three-year-old boy, it both sparked my imagination and inspired me to learn to read. (Play it yourself by clicking here.)
The staggering takeaway is that I was able to play Zork on a TRS-80 running on only 4 kilobytes of Random Access Memory (RAM). In contrast, today’s Apple computers are offering 8 gigabytes of RAM, or 8 million times more than the TRS-80. The phones in our pockets would be considered supercomputers just a few decades ago.
Solace can be found in this rate of technological progress, especially at a time when so much in the world seems to be regressing. Faster, more powerful computers will be able to bring new understanding to weather forecasting and climate change, medical diagnosis, robotics, and may even help us design new systems of governance and economics.
We must sustain this momentum. A positive future depends upon it.
Hope for continued advancement in computing power, and in particular, supercomputers, can be found in the pressure cooker of international relations. Competition among nation-states has long been a powerful catalyst for progress. Much as the Soviet satellite ‘Sputnik’ zipping across the sky helped motivate a new generation of scientists, the national quest to build the world’s most powerful supercomputer is accelerating demand for science, technology, engineering and math (STEM) research.
Japan recently began soliciting bids to build the world’s fastest supercomputer by 2018. This new machine would have a processing power of 130 quadrillion calculations per second, or 130 petaflops, and would cause Japan to leapfrog past China – the current world record holder for the world’s fastest supercomputer at 93 petaflops. At a cost of US$173 million, the new supercomputer would temporarily put Japan in first place. However, the U.S. has indicated that it will attempt to reclaim the title in 2018 by unveiling a machine with a speed of 200 petaflops.
While the rise of anti-science and anti-intellectual forces in the United States seems to portent a slowing of technological progress, we can take some comfort in knowing that the rest of the world will only accelerate their basic scientific research to gain an advantage in a multi-polar world. These technological advances will diffuse across national borders, raising the wealth tide for most ships of state.
Progress is made under pressure. A supercomputer arms race between nation states may be just what we need for continued advancement and prosperity. Bring on the exaflops.