Quantum computing explained: D-Wave on NASA's system
- By John Breeden II
- Sep 10, 2013
If one of the 35 million books in the Library of Congress has a big red X on the inside cover, how long would it take a man to find it? Opening each book might take the searcher hundreds of years, unless he was very lucky.
But what if the searcher could replicate himself into 35 million different people, each one existing in a parallel universe? All 35 million people would head to the library and look inside a different book. If no X were discovered in an individual book, that searcher would simply disappear, until the last man standing would be holding the correct book. The same impossible problem could be solved in a few minutes.
That’s one way to illustrate the difference between traditional and quantum computing, and it explains why the government is so interested in machines that can virtually try all possible solutions at once and find the best answer more quickly. That’s also why NASA, Google and the Universities Space Research Association have formed the Quantum Artificial Intelligence Lab, to explore quantum computers’ potential to tackle problems that are too difficult or perhaps impossible for supercomputers to handle.
In fact, some problems can never be solved by traditional computers, according to Eric Ladizinsky, co-founder and chief scientist for D-Wave Systems, which built a quantum computer for NASA and is working on even more powerful versions that could one day soon crack open the mysteries of the universe.
With traditional computers, the circuits are either on or off, and the binary code is represented by ones and zeros. Adding more processors increases the computer’s power linearly. By contrast, a quantum computer uses quantum bits, or qubits, the quantum equivalent of a traditional bit. Its circuits exist in all possible states at the same time, – a one, a zero and whatever is in between – and this superposition vastly increases the potential processing power.
The National Science Foundation recently posted an animation in which theoretical physicists John Preskill and Spiros Michalakis explain the principles of quantum computing. Superposition becomes useful when quantum bits work together, multiplying the ways they can be correlated. The correlations are richer, and that richness increases markedly as even a few hundred qubits are added — so much so that these correlations couldn’t be described with classical bits. “You’d have to write down more numbers than the number of atoms in the visible universe,” the scientists said.
But that random richness requires that the calculations be run in a stable environment completely isolated from the outside world because observation would destroy the delicate random superpositions. Likewise, there can’t be any leakage of information from the quantum computer to the outside world. That decoherence, or what the scientists call “the big enemy,” would destroy the quantum calculations as well. Only now, they conclude, “are we developing the technological capability to scale-up quantum systems.”
That’s where D-Wave comes in. The D-Wave quantum computer takes a ring of metal and cools it down close to absolute zero. Then other factors are eliminated to combat the decoherence that can destroy the quantum calculations. Light is removed by sitting the machine inside a black box. Radiation is shielded, and sound is reduced as much as possible. All air is also removed from the enclosure. The result is that when a current is applied to the ring, scientists can measure the superposition – 100 percent of the current is going clockwise at the same time that 100 percent of the current is going counterclockwise. That dual state is harnessed to solve problems.
The secret to D-Wave's approach versus other quantum computing companies is that it has been able to achieve quantum phenomenon using concrete parts. That means D-Wave can build its quantum computers in a more traditional way, as opposed to trying to work with atoms and electrons directly.
Ladizinsky says much of the government’s interest in quantum computing has to do with code breaking. To break 128-bit RSA encryption the traditional way would take 2,000 workstations and supercomputers about eight months. For 256-bit encryption, it’s a million years. And for 600-bit encryption, it would take the age of the universe. But with quantum computing, the size of the problem doesn't matter so much, because a powerful enough quantum machine could look at all the possibilities at once. Although Ladizinsky says the D-wave machine is not specifically designed to break encryption, he knows others are experimenting heavily in that field.
Ladizinsky would like to use his company's quantum technology to solve climate change, fight diseases and further biological research. NASA is also is interested in using its 512-qubit machine to study machine learning and artificial intelligence, he said.
Researchers are working to improve communication among qubits. Right now, each qubit can talk with the six others sitting around it, but in the future Ladizinsky said he would like to see the entire matrix communicate. And unlike traditional computers whose processing power grows linearly, the quantum machines’ power grow exponentially, blowing Moore's Law out of the water. Moving from a 128-qubit computer to today's 512-qubit machine increased processing power by 300,000 times, Ladizinsky said.
As even more power is added to D-Wave's quantum computers, those seemingly impossible questions of yesteryear may begin to fall in range of today's scientists. "In a sense, we are harvesting the parallel worlds to solve problems in this one," Ladizinsky said.
John Breeden II is a freelance technology writer for GCN.