 |
||
| Home | Forums | Article Search | Subscribe & Shop | Contact Us | Log Out |
| Quantum Computing |
 Email This
 Print ThisView My Personal Library |
|
Hard Hat Area February 2005 • Vol.5 Issue 2 Page(s) 46-49 in print issue |
|
|
Quantum Computing The Power Of Qubits |
|
The physical tools used in computation have changed steadily throughout history. From slash marks in the mud to the abacus and beyond, humans have developed thousands of ways to help them perform computations. (As an unfortunate consequence, humans also have developed thousands of ways to make errors while performing computations, but that's a discussion for a different story.) Even within the history of computers, the physical tools have changed dramatically. The earliest computers made use of gears and vacuum tubes. Now, obviously, miniscule logic gates, transistors, and wires on silicon chips can perform calculations with amazing speed and accuracy. It might be hard to fathom now, but at some point, today's computing tools will become obsolete. What exact technology will replace them is uncertain, but scientists and researchers are excited about the possibilities of quantum computing, which promises to be exponentially more powerful than today's classic digital computers. However, most computer scientists think quantum computing only will be beneficial for researchers and scientists performing overwhelmingly complex calculations.  The Basics Quantum computing incorporates the idea of processing data by making use of physical phenomena related to quantum mechanics. Equations spelled out in quantum mechanics give you the ability to predict what quantum computers will do. Quantum mechanics is the most important theory for many fields of physics, formulating nearly all other theories. One of the features of quantum mechanics, called quantum interference, spells out the theoretical power behind quantum computers. The Centre for Quantum Computation in England says although it's difficult to pin down the structure of a quantum computer, it will be somewhat similar to a large network of today's computers, performing computations that are different but related. All the parts of the quantum computer affect each other through quantum interference. Quantum vs. Classic Digital computer data traditionally is built on the power of a bit. All data is represented by a series of zeroes and ones. The bits follow all laws of classical physics. In quantum computing, the individual bits, called quantum bits, or qubits, follow the laws set forth in quantum mechanics, especially quantum interference. A qubit following quantum mechanics is vastly different from a traditional binary bit in a digital computer. The electronic state of the qubit will determine whether it's a zero or one. One complication: Quantum mechanics says the qubit can have an electronic state that's both a zero and a one (called a superposition). You can think of superposition as a blend between the zero and one found in traditional computing bits. This complication is beneficial, though, because it's also the power behind quantum computing. In a traditional computer, a set of two bits will yield four unique configurations (00, 01, 10, and 11). However, the two bits can only store one of those four configurations at a time. You can calculate the number of configurations for any set of bits. Just multiply 2 to the nth power, with n representing the number of bits you're using. (The "2" is representative of the two possible positions of the bit, a zero or a one.) If you're using a set of four bits, you'll have 2 to the 4th power, or 16, configurations available. In simple terminology, a qubit, however, can store multiple configurations at the same time, giving it far more power than a traditional computer bit. That power grows exponentially when you use a set of multiple qubits. Referencing the above example, a set of two qubits could store all four unique configurations at one time. As you increase the number of qubits in use, the number of configurations it can store at once increases exponentially. To calculate the number of configurations a set of qubits can store at once, multiply 2 to the nth power, with n representing the number of qubits in the set. In other words, a set of four qubits could store 16 different configurations at one time, but a set of four digital bits could store only one of its possible 16 configurations at one time. The power of quantum superposition gives quantum computing exponentially more computing power than a traditional computer. Not only can a set of qubits store multiple configurations at once, it can perform operations on all of the configurations at one time. Because quantum computers enjoy exponentially more power than traditional computers, traditional computers cannot match them, even with more memory and processing power. One example, published in a Caltech paper in 2000, showed that a quantum computer containing 500 qubits would represent a superposition equal to as many as 2 to the 500th power states. The equivalent in a traditional computer would be a string of 500 bits (zeroes and ones) for each state represented in the quantum computer. To match the computational power of the 500 qubits, a traditional computer would need a number of CPUs equal to about 10 to the 150th power (based on 2000 technology). For comparison, some estimates say the number of atoms on earth equals about 10 to the 50th power. Quantum Strengths When quantum computing will become a reality is unknown. Once a quantum computer is built, scientists and researchers undoubtedly will develop several new applications that need and can harness the quantum computer's amazing power. However, many scientists have already theorized about some of the tasks in which quantum computers will excel. Cryptography. Cryptography is the method of converting readable information into unreadable information via a secret coding format. By deciphering the code, the user can turn the unreadable data back to a readable format. Cryptography is the area of quantum computing that the nonscientific community probably will encounter first. "Using the ideas of quantum information processing to establish a cryptographic system that's undefeatable is likely," says David J. Wineland, who is leading a group of scientists making impressive strides in quantum computing research. "It's a reasonable guess that one of the first applications of ideas of quantum information processing would be for cryptography." By using quantum computing power to encrypt computing data, the code could become so complex that it would be essentially unbreakable. Factorizing large numbers. After Peter Shor of AT&T's Bell Laboratories developed the first quantum algorithm in 1994, the idea of factorizing large numbers efficiently became realistic. Shor's algorithm also gave quantum computing its first useful and unique function; traditional computers cannot factorize large numbers efficiently. Although the idea of quantum computing had existed for several years before Shor's algorithm, the algorithm provided the first killer application that required quantum computing. As with any quantum computer algorithm, Shor's algorithm can only offer a high probability that it has calculated the correct answer. If you run the algorithm again, the probability of a correct answer increases even more. Simulating quantum-mechanical systems. These simulations are too complex for traditional computing systems. Because of the exponential computation power of qubits, though, they easily would have enough power to perform these complex simulations. (Such simulations are complex mathematical operations used to predict the behaviors of microscopic particles.) "These [simulations] probably will be realized before big factoring problems are realized," Wineland says. "People are hopeful about that." Breakthroughs Recent scientific advances have made the possibilities of quantum computing more realistic. Two groups of scientists working independently both announced key findings in Nature magazine in mid-2004. Both groups transmitted characteristics between atoms, which is at the heart of quantum computing. (Previous experiments involved the transmitting of quantum characteristics of beams of light.) Although the characteristics moved only a tiny fraction of an inch, that distance is typical of what would be required in a quantum computer. One atom's complex set of traits, called its quantum state, was transmitted to a second atom in the experiments using a phenomenon called entanglement. Albert Einstein initially proposed the idea of entanglement, which allows researchers to create a relationship between two atoms or particles. Once the first atom takes on a set of properties, the related atom automatically takes on the same set of properties, even though there is no apparent physical connection between the atoms. The recent breakthroughs highlighted quantum teleportation, which will be an important factor in quantum computing. Because no physical connection is needed between the qubits, they can interact without limitations related to location. The group of scientists at the University of Innsbruck in Austria performed the experiment using calcium ions, while the group at NIST (National Institute of Standards and Technology) in Boulder, Colo., made use of beryllium ions. Wineland, who is the group leader of Ion Storage at NIST, headed the group in Boulder. "We were in a friendly competition," Wineland says of the two groups. "It would be natural for both groups to do the same things. We just decided not to kill ourselves off competing and to work together at a better pace." Wineland became involved in quantum computing experiments in the early 1990s when he was trying to use entanglement while working with improving atomic clocks and using spectroscopy. After seeing Shor's algorithm, Wineland decided the problems he was working with were similar to the problems in developing a quantum computer. Overcoming Problems As with any type of computing system, developing a good system for error correction will be important to the eventual success of quantum computing. Error rates in the early experiments have been of concern to the scientists, but Wineland says they aren't debilitating to the overall process. "The errors are caused by things we know in principle that we can make better," Wineland says. "It's not easy, but we're optimistic we can get better and faster. . . . Most of us are beginning to feel now there is no evidence, there are no fundamental problems, that we can't do this." One problem that leads to errors involves fluctuations within the laser beam that manipulates the qubits. However, steady improvements in laser technology should alleviate those problems. Another problem deals with decoherence, which is the natural tendency of a qubit to change from one quantum state to another as it naturally interacts with the environment. Researchers continue to work at combating decoherence, and they've made significant strides, providing hope that it won't be an issue when quantum computers become a reality. "Now we're in the stage of the game where to be serious, we need to be able to conceive scaling it up," Wineland says. "There is a path where we can see how to do it. At this stage, though, it's not clear who will be able to solve these technical problems." The Future "Star Trek" fans may hear the word teleportation used in reference to quantum computing and have dreams of quantum computing advances leading to Scotty beaming them up sometime in the near future. Wineland, however, says the recent findings have nothing to do with teleporting matter. "We're transporting information," he says. "In this case, quantum information." The experiments are not going to yield a working quantum computer in the next few years, but Wineland says they represent another step forward. "The one thing that's interesting . . . is very probably this is the way you'll transmit data in a quantum computer," Wineland says. "Admittedly, this is just one piece off a large puzzle, but it's a tool that will go in the toolbox." For scientists and researchers, quantum computing will give them an invaluable tool for performing calculations that essentially are impossible with today's computational technology. For everyday computer users, the eventual benefits of quantum computing aren't as easy to decipher. Chances are very slim that a typical PC user will ever need a fraction of the computing power that quantum computing would offer. And, because current quantum computing theories indicate a classic digital computer would run in conjunction with the quantum computer, today's PC technology doesn't appear to be fading away anytime soon. Still, no one will know exactly what the uses for and benefits of a quantum computer will be until a working unit appears. And no one knows exactly when that will be. Even making educated guesses as to a timeline for the technology is difficult. "Everybody is thinking correctly that this is way out there in the distance," Wineland says. "But what is encouraging is where the technology is headed. . . . I would believe someday we'll have quantum computers—but it may be beyond where I'll see it. "You could imagine in 10 years we could identify a breakthrough technology," Wineland says. "It may not be what people are thinking about now. It's not fully known what will happen. Ideas are still kind of percolating."  by Kyle Schurman View the graphics that accompany this article. (NOTE: These pages are PDF (Portable Document Format) files. You will need Adobe Acrobat to view these pages. Download Adobe Acrobat Reader)
|
