Are there limits to the development of the computer industry? Taking into account the pace of this development, over the past half-century, there has been an irresistible impression of the prospects for further improvement of information technologies. Although no further than in 2007, the notorious Gordon Moore stated that his law, which predicts a doubling of the number of transistors on an integrated circuit chip every two years, appears to soon cease to operate in the trivial reason of the atomic nature of matter and the limitation of the speed of light …  We will not argue about the possibilities for overcoming the above-mentioned speed, but it is even possible with regard to atoms that the situation is far from so sad. After all, as in the mid-1980s, when personal computers were just starting their triumphant way to replace mainframes, the more famous American physicist, Nobel laureate Richard Feynman (1918-1988), argued: "It seems that the laws of physics do not represent any limit to the reduction in size computers up to the point that bits become the size of atoms and will start to dominate quantum behavior. "
Actually, quantum computers that partially overcome the limitations of the atomic nature of matter and which were described in detail by ITC.UA s even a decade ago, there on the paper has been years since the thirty-plus, thanks to the research of the same Feynman proposed the idea of such a device in 1981 – although even a little earlier and independently of him did Simferopol-born mathematician Yuri Manin. Recall that in "ordinary" computers, media, from punched cards and transistors, contain information encoded in a binary number system: the presence or absence of a hole, the induction of the magnetic field is greater or less than the threshold value and simply the state "incl. / off "are translated into bits consisting of zeros and ones.
The internal view of the "quantum computer" from IBM.
Quantum counterparts of "bits", called "qubits" (from the quantum bit – which, by the way, in many languages successfully coincides with the ancient Greek and the ancient Roman unit of length measurement, the analogue of the "elbow"), as the media of a computer of a qualitatively different type, are distinguished by the ability for the so-called "superposition" – the ability to be simultaneously in both states (conditionally speaking, "1" and "0") simultaneously, but only until such a state is measured. In theory, such a strange feature of subatomic particles has been the subject of many disputes and complex hypothetical situations, like the famous (and very unfortunate) "Schrodinger cat" for almost a century, forced to be both alive and dead in a box with a radioactive substance in which a possible decay one of the atoms causes activation of the deadly poison. However, in practice, such obscure properties of atoms have long been used – and the field of computer technology is no exception.
The trick is that if at the very beginning of the calculations to translate a system consisting of quantum media with input data into a state of superposition, such calculations will be performed for the entire set of data received in parallel – that is, with a huge acceleration in solving the problem. True, there arises the problem of measuring such computations – because, like the Schrodinger cat, if you open the box and "look" at its state, it will always be either live or dead, and "qubits" when measuring their data will be able to give we have only one answer, in spite of all the "parallelism" of the computations preceding this measurement. So, with all the advantages and unprecedented advantages of this kind of computer, it will not be possible to use it for any calculations.
In the 1990s, several possible schemes for the operation of a quantum computer were proposed, named after the scientists who nominated them. So, the algorithm of Peter Shore from Bell Laboratories provides that we may be interested not in the whole sequence of values of a function, but only in its period, which is much more accessible for measurement. But with the help of this algorithm on a quantum computer can be at an unprecedented speed – somewhere in 100 million times faster! – solve the factorization problem, that is, the definition of simple multipliers of large numbers, which, in turn, allows you to almost immediately decrypt cryptographic algorithms with a public key, because the existing RSA cryptosystems are built on the inaccessibility of this task to the current capacities of conventional computers.
An element of a quantum computer in the artist's view: a nanotube with fullerenes – molecular compounds in the form of a soccer ball and with nitrogen atoms inside, which act as "qubits". © 2013 Karl Nyman; OxfordQuantum.org.
It turns out that quantum computers can not in any way claim the place of ordinary, electronic, they are able to act as a supplement to them, organizing assistance in solving a particular kind of tasks. So the prospect of soon acquiring pocket quantum PCs for home users is still very remote – especially as there are many obstacles to the simple implementation of the idea of computers based on qubits. First, in order to "start" the solution of any problem on a quantum computer, it is first necessary to produce, as they say, "initialization" of qubits, leading them to a "zero" initial state – and for this, in turn, requires the cooling of information carriers to temperatures , close to absolute zero. So, devices of a new type can exist only in special cryocameras with extreme freezing.
Secondly, – and this problem is much more complicated, – quantum bits, – are just as prone to errors in calculations as ordinary computers. In the mid-1990s, the level of error reached an extra 10 percent – whereas the theoretically acceptable value is 0.0001 percent. At present, scientists have managed to reduce this figure to less than one percent, and although there is still a lot of work to be done, according to moderately optimistic forecasts, it is planned to achieve satisfactory performance of quantum computers sometime in the early 2020.
An IBM employee inspects the cryostat with a new prototype of a commercial quantum processor inside.
These errors occur because of a phenomenon called "decoherence" – or, more simply, loss of communication between two interacting particles. And this is just another unique feature of subatomic elements, which can bring a lot of usefulness when using qubits as a carrier of information: quantum particles, even being very well isolated from each other, can be in a connected (or, in the expression of all the same Schrodinger, "Confused" – entangled) state, in which they somehow depend on each other. In other words, the quantum state of one particle can not be described separately from the other (for example, if the "helicity" index of the first of them turns out to be positive, then the second one will necessarily be negative), which means that measuring one of the particles will also mean an instantaneous termination "Uncertainty" about its pair.
Theoretically, this "confusion" is preserved at any distance: this phenomenon of transferring a quantum state from one particle to another has been called "quantum teleportation" (which is has little in common with teleportation in the sense of everyday – fantastic – word usage). The latest relevant experiments in this area were carried out by Chinese scientists in June 2017: a phenomenon of conservation of the interaction of bound quantum particles at a record distance of 1203 km was demonstrated with the help of a special satellite transmitting photons of infrared light. So even if there is any hidden interaction between the "members of the pair", then its speed should many times exceed the same speed of light, which – as is commonly believed – impose an insurmountable restriction, including the further development of computer technology! However, as R. Feynman repeatedly mentioned, he said: "I think it's safe to say that no one understands quantum mechanics … If possible, do not ask the question:" But how can it be? "- because you "Suck" into such a dead end from which no one has ever been selected. "
" Mo-tzu "is the world's first satellite intended for the quantum transmission of information over a communication channel, guaranteed to be protected from hackers. In June 2017, he was able to broadcast "tangled" photons at stations located in the Chinese cities of Dalinh and Lijiang, the physical distance between them is 1203 km.
The head of the Chinese group of scientists, a specialist in quantum "connectivity" Jian-Wei Pan is very optimistic: according to his assurances, several more satellites of this kind will be launched in the next five years, and by 2030 the quantum communication will become international, so that it will be possible to talk about the real "quantum Internet". Most other scientists are more cautious in their conclusions – for example, the Canadian physicist Thomas Jennevein believes that the very concept of "quantum Internet" may sound nice, but so far it remains not very specific, and the technology itself is still in its infancy: to control quantum Signals on a level sufficient for the transfer of information scientists at the moment are not able. "Mo-tzu" worth a hundred million dollars is only able to broadcast and receive beams of quanta, but do not store information – and since quantum signals, unlike conventional electronic ones, are not subject to amplification, the network of satellites only their radical cheaper, but also the development of non-existing repeaters – and it is by no means a fact that today's living generations can see all these hypothetical achievements.
Actually, the "quantum Internet" is unlikely to bring smth Bo is fundamentally new in the transfer of information through the channels of remote communication: for communication between people ordinary "ones" and "zeros" is quite enough, and the involvement of superposition and entanglement nothing to it will not add. After all, since any measurement of a quantum system changes its state, "quantum information" can not be copied in the traditional way; another thing is that just as quantum computers are a potentially useful addition to electronic computing devices designed for high-speed performance of specific tasks, so in the field of communications, the "quantum Internet" can become a specialized version of the "ordinary" one. And above all – more secure and secure, because not only decoding, but also encoding messages turns out to be the strong side of quanta.
Photograph of a crystal with "confused" photons. © Félix Bussières; University of Geneva.
Arthur Eckert, a British scientist (in the good sense of the word combination) of Polish descent, an expert on both quantum physics and cryptography, who demonstrated the idea of such use of quantum computers in the field of communication in 1991, on paper, exactly how the phenomenon of quantum confusion can serve to achieve an unprecedented level of security in the encryption of messages. It is, in particular, that one of the two coupled photons is transmitted over a long distance, where it interacts with the third particle, and the state of this third photon is transmitted not only to the second one, with which it is directly encountered, but also instantaneously "teleports" to its the twin, the photon number 1. Due to this property, the transmission of secret messages is achieved: when two people exchange a "confused" pair of particles, a quantum state – a kind of "information", even if not in the full sense of the word – is transmitted between them without any kind of material carrier and direct interaction, which literally excludes the possibility of interception of such information by a third party.
Not so simple to understand the scheme of the "quantum Internet" device.
Currently, several companies already offer commercially available devices based on the use of this kind of cryptography. For example, the Swiss idQuantique back in 2007 provided encoding technology for the transfer of voting results in elections from Geneva to Bern, announcing that complete security and protection of the broadcast data is guaranteed by the iron laws of physics. True, those features of the behavior of quantum particles, thanks to which it is impossible to "spy" their state, since this will mean a change in that state, and hence the fact of "interception" will immediately become known – these features make it practically meaningless to transmit information in such a "self-destroying" but they allow you to apply an excellent way of exchanging keys to decrypt encoded information transmitted in the usual way. This technology is called the "quantum key distribution": after the keys are transmitted and confirmed – and therefore no one could "spy" them along the way, otherwise the change in the state of the particle would be obvious – you can start encrypting the information and its immediate translation, which, thus, will be associated with a minimum of possible risks.
Of course, the currently practiced version of quantum technology is not deprived of a number of shortcomings: hackers even reported in their time about successful hacking "extremely safe th "communication channel, although scientists also assure that all possible leaks were associated with technical errors in the implementation of theoretically invulnerable quantum coding (which they are constantly eliminated as technology improves). Another problem is somewhat less pleasant: the current version of the commercial implementation of quantum key distribution requires expensive server equipment and an optical cable, this is transmitted a thousand times or even ten thousand times slower than usual information and no further than a distance of 100 km. So it remains only to wait for the hypothetical introduction of quantum repeaters, so if not to broadcast the information, then transfer the keys for decryption with the help of satellites like "Mo-Tzu."
Predicted use of quantum satellites Quantum key distribution.
As can be seen from the result of this small survey, projects based on the laws of quantum mechanics – at least for now – are far from representing a new revolution in the field of information technology. Quantum computers and even more "quantum Internet", even in theory and on paper, are not only replacing, but complementing the currently existing methods of computing and transmitting data. The former are too complicated to maintain specialized, designed to solve certain tasks, and therefore are more suitable for "cloud" computing, already practiced by some "pioneers" – and the latter are potentially able to help achieve the highest level of security of information broadcast over the network, but require still a considerable improvement. Nevertheless, as reported in its latest victorious reports of IBM, its project of those same cloud computing on a relatively small quantum computer has already benefited more than 60,000 users who have spent 1.7 million or more quantum experiments. So, it is likely that the new discoveries are not far off.