Issue link: http://cp.revolio.com/i/324708
THE DATA CENTER JOURNAL | 11 www.datacenterjournal.com QuantuM CoMputing a lthough it doesn't focus on a material per se, quantum computing has for years been the heir apparent to traditional electronic com- puting using silicon. Rather than simply employing a different atomic structure and composition, quantum computing relies almost entirely different physical prin- ciples: particularly, quantum mechanics. eoretically, a quantum computer could run laps around even the best silicon- based machines thanks to the properties of "qubits": quantum bits, which are able to hold both 1 and 0 (on and off ) states simultaneously. Bits in electronic devices are either 1 or 0 exclusively. Although many pundits view the arrival of quantum computing as simply a matter of time, this technology faces a number of practical and economic chal- lenges. In particular, maintaining us- able qubits typically requires equipment temperatures very close to absolute zero (about –273°C), requiring expensive cryogenic infrastructure. (By contrast, your average air conditioner—albeit on a larger scale—is enough to keep most data centers sufficiently cool to function.) Assuming quantum computing comes to fruition, the most important matter won't be how much computing per- formance it can deliver. e critical con- sideration will be how much computing power per dollar it can deliver—and that dollar includes infrastructure, electricity, capital, space and other costs. A quantum computer offering the power of a thou- sand servers is economically infeasible if it costs more to buy and run than that same number of servers. Currently, computer-technology company D-Wave claims to offer quan- tum computers—specifically, a kind of quantum computer that employs so-called quantum annealing. Debates as to whether these machines truly deliver any kind of quantum computing (as well as whether they can even exceed the capabilities of "classical" computers) are ongoing. e company's latest model has been described as being the size of a sauna, with most of that bulk being dedicated to keeping the chip well within 1°C of absolute zero (for comparison, liquid helium is about –269°C). e chip itself is roughly the size of a thumbnail. Like the flying car, quantum comput- ing may become one of those technolo- gies that is technically possible but simply impractical, relegating it to popular fiction and a few prototypes but no real commer- cial presence. graphene Graphene boasts a number of proper- ties that make it attractive for application in a wide range of fields, from computing to water filtration and more. It is a single- atom-thick layer of pure carbon forming a hexagonal lattice. Far from being exotic, graphene is present in graphite—the Rus- sian-born scientists who discovered it did so through a repeated process of peeling away layers of graphite using Scotch tape. It is extremely strong—reportedly, about 100 times the strength of steel, pound for pound—as well as an outstanding thermal conductor. Its atomic configuration also makes it a good electrical conductor. e material's unique and numer- ous remarkable characteristics have drawn extensive research effort (and dollars) in a wide array of fields. Naturally, however, one is tempted to question whether a material with the apparent potential to do it all will actually end up doing any one thing well. But such questions won't stop researchers from trying. IBM, for instance, designed a graphene transistor with a gate length of 40 nanometers and a cutoff frequency of 155GHz. But before you start salivating for a new graphene PC chip that blows Intel's products out of the water, note that this transistor is not yet suitable for digital electronics because of graphene's lack of an energy gap—one of the properties of silicon. us, although it has potential for analog electronics, digital applications have longer to wait. e main factor in graphene's potential—whether for computing or other fields—is whether it offers the right combination of properties. Presently, it's a solution in search of a problem, and that status raises doubts about its future, at least in the near term. CarBon nanotuBes Imagine a sheet of graphene rolled into a tube—that's basically what a carbon nanotube is. Like their close cousin, carbon nanotubes offer a number of unique properties, and researchers are pursuing their use in computers as a silicon replace- ment. Engineers at Stanford University The remarkable advancement of the semiconductor industry over recent decades, described by the much-celebrated Moore's Law, may finally be slowing as silicon manufacturers struggle to bring their next-generation process technologies to production. In an effort to keep innovation going, researchers are investigating all manner of materials as potential successors to silicon. Doubtless, some candidates for this vaunted position will fall by the wayside for either technical or economic reasons; the question is whether any will survive the rigors of investigation and yield a product-worthy result. So, all aboard the starship Enterprise for a look at what might—or might not—be the computing material of the future.