Quantum computer: from dream to reality
Now the very abstract ideas underlying quantum physics are being translated into reality thanks to new technological capabilities in the field of nanotechnology and optical interactions. One of these ideas, the idea of a quantum computer, will be discussed further. I will try as much as possible more popular.
To be consistent, it is impossible not to mention the theoretical achievements of quantum physics, which made it possible in practice to create such devices.
In general, what is quantum mechanics? It is known that quantum mechanics is a theory by which microscopic systems obey, for example, atoms. The microworld - atoms, electrons, photons and other particles - lives according to special laws. There, not just everything is very small, everything is completely different there, and many phenomena of the microworld have no analogues in the usual macrocosm, which makes them seem fantastic.
A characteristic feature of quantum mechanics, which is imprinted in the name - in some physical systems, energy is quantized , that is , it can be equal to one of a certain discrete set of numbers, or even any other value (forbidden zone). If you try to imagine such a situation in the macrocosm, you can imagine, for example, that objects have a temperature that is expressed only by an integer number of degrees. That is 10, 20, 31, 36 degrees - maybe, but 36.6 - is simply impossible. It is possible to heat and cool items, but at the same time the temperature will jump back and forth immediately to a degree.
')
In particular, the energy of the electromagnetic field is emitted only in the form of discrete indivisible portions. Such a portion is called a quantum . Max Planck made the assumption of the existence of quanta in 1900–1901, thereby laying the foundation for quantum theory, quantum mechanics, and much more with quantum adjectives - including computers.
Another amazing property of photons, electrons and other particles is that they can exhibit the properties of both particles and waves . Light presented in classical physics as an electromagnetic wave, turned out in quantum physics as a set of photons, like particles. But conversely, particles, such as electrons, exhibit properties characteristic of a wave: they interfere with each other and diffract on obstacles (that is, bend around them). This property of particles gave the basis in the early years to call the theory describing them wave mechanics, and only later the name "quantum mechanics" was established.
And finally, the third fundamental property: a quantum object, in contrast to the classical one, is initially statistical . It is possible to describe the location of a particle not directly observable in a certain place only with a certain probability.
There is also a significant limitation on observations and measurements in the microworld: the Heisenberg uncertainty principle : it is impossible to measure simultaneously the speed and coordinates of particles.
As long as we do not measure objects, they behave even worse. If a quantum system can be in several states and it is not known in which it is located, then we are talking about a superposition of states . It can be said that it is unknown what state the system is in, or that it is in several states at the same time, it is a matter of interpretation. In any case, when measuring, the system selects one of the states.
A well-known illustrative example is a thought experiment called “Schrödinger's cat”: a live cat, a container with a poisonous gas and a radioactive core are placed in a closed box. If the nucleus disintegrates, it activates a mechanism that opens the gas tank and thereby kills the cat. The probability that the core will disintegrate in an hour is 50 percent. An hour later, the cat in the box is alive with a probability of 50 percent. From the point of view of quantum mechanics, while the box is closed, the cat is in a superposition of two states (either alive or dead; and alive, and dead; neither alive nor dead - however you please). At that moment, when the observer opens the box, he sees whether the cat is alive or dead.
Finally, another important phenomenon for us is quantum entanglement, which is also confusion, cohesion, and sometimes coherence. They say about entanglement when the state of two (or more) quantum systems must be described in conjunction with each other, even if the systems themselves are spaced apart. Accordingly, the physical properties of each of the systems are associated with the physical properties of the other, while they may not be close by and not be connected in any way.
If two entangled systems are in a superposition of states, then by measuring the state of one, you can find out the state of the other . For example, you can confuse two atoms, the spin (a certain quantum characteristic) of one of which will be directed up, and the other down, and we will not know which atom has which spin. But by measuring the spin of one atom, we immediately recognize the spin of another, even if they are spaced apart.
All these and many other features of the microworld make it possible to build a quantum computer.
A quantum computer, like any other computing device, must operate with numbers. The simplest device capable of representing numbers is a device that can be in two stable states. So, conductors with a current can be in two stable states: when there is no current, it corresponds to a value of 0, and when there is a current, it corresponds to a value of 1. There should be no other steady state. The use of such devices for conducting operations on numbers is possible due to the binary recording of numbers — information such as the choice between two options (digits 0 and 1 or, equivalently, two system states) is stored in such a cell. It is said that the amount of information stored in a binary cell is 1 bit. Two binary cells store 2 bits of information. This corresponds to the choice between 2-bit binary numbers 00, 01, 10 and 11. In a register consisting of N binary cells, N bits of information are stored, which corresponds to the choice of one N-digit binary number of all possible (all such numbers are 2 ^ N ).
Now let's see how the storage of information in a quantum computer is different. The information itself can also be represented in the form of binary numbers, that is, chains of zeros and ones. One digit of a binary number is stored in a binary cell. This cell is called a qubit and is a quantum system, one of the states of which corresponds to the digit 0, and the second - to the digit 1. We denote these states as ψ1 and 2. There are still no differences from the classic computer. But it is at this point that a significant difference appears. The fact is that for quantum systems the so-called superposition principle holds: if a quantum system can be in one of two states, say, in the state ψ1 and in the state ψ2 , then it can also be in the whole family of states that are constructed as linear combinations of the original. This means that the initial states are multiplied by some (complex) numbers and added together. If the numbers c1 and c2 are chosen, then the state is c1ψ1 + c2ψ2 .
In particular, if a qubit is brought into the state (∣0〉 + ∣1〉) / √2 , then this means that both the digit 0 and the digit 1 are simultaneously written in it.
So, in one binary cell of a quantum computer, called a qubit, can be stored not only one of the two digits of the binary number, 0 or 1 (as it would be in the case of a classical computer), but both of these numbers at the same time. For example, 4 binary numbers 00, 01, 10 and 11 can be stored in two qubits simultaneously. And if some register of a quantum computer contains N qubits, then such a register can simultaneously store 2 ^ N binary numbers of length N. And when a quantum computer acts all these numbers are processed simultaneously. This is quantum parallelism. If we had only classical computers at our disposal, each of which works with binary numbers of length N, then 2 ^ N computers would be necessary for simultaneous processing of 2 ^ N such numbers. If we managed to build a quantum computer containing N qubits, then one (!!!) this computer simultaneously processes all 2 ^ N numbers.
Unlike classical computers , quantum computers are not universal : there is no algorithm for solving computational problems on a quantum computer for any computing task. So far, only a small number of quantum algorithms have been found.
The most famous quantum algorithms:
1) Shor Algorithm (factorization)
2) Grover's algorithm (quick search in an unordered database)
3) Algorithm Deutsch - Jos (answer to the question, constant or balanced function)
Due to the use of the quantum entanglement phenomenon and the principle of superposition, quantum algorithms have a significant increase in execution speed compared to the corresponding classical algorithms.
In a bit more detail, let's stop on Shor's algorithm, which allows to solve any of two mathematically equivalent problems:
1) find the period of a complex periodic function or
2) decompose a very large number into prime factors
The second of these tasks is of significant practical importance because it is used in cryptography. The fact is that in one of the cryptographic methods for encrypting and decrypting secret messages (public key encryption) they use large numbers for which their factorization is known. It is clear that such numbers are easy to obtain: it is enough to multiply a large number of prime numbers, and we get a very large number for which the factorization is known. The recipient of the encoded secret message can decode it because the decoding procedure uses factorization of a long number, and he knows this decomposition.
If the enemy could factor this number into prime factors, he would also be able to decode the message. However, this decomposition takes a lot of time. Therefore, from a practical point of view, it is impossible to decode such a message. But if the enemy had a quantum computer (of sufficient power, that is, working with sufficiently large numbers), then he could decompose long numbers into simple factors and therefore could easily decipher such messages. This common cryptography method would stop working. This is one of the arguments that make the creation of a quantum computer important.
Physical:
Mathematical:
These problems can be described in one word - scalability . Attaching additional memory to a regular computer is a simple routine procedure. Attaching each new qubit to QC is for now a piece work.
The phenomenon of decoherence is that the system being measured loses its specific quantum properties. In other words, the pure state quickly turns into a mixture when the quantum system interacts with the environment.
To combat decoherence, various methods of isolating a quantum system are developed, on the one hand, including the use of extremely low temperatures and high vacuum, and on the other, an introduction to quantum computing of codes that are resistant to decoherence errors (usually in such circuits encoded by the state of several connected physical qubits).
The accuracy of unitary operators, according to various estimates, should be from 99.99 to 99.999999 (depending on the specific implementation).
Ionic traps are the first and most well-developed idea. One of the superconducting computers will be discussed further. Well, quantum dots are considered the most promising option.
In February 2007, the Canadian company D-Wave Systems assembled a 16-qubit quantum computer, which founder and CEO of Georgy Rose called the most powerful quantum computer ever built, and the first one that can run commercially-significant applications. In November 2007, at the conference dedicated to supercomputers, the same company D-Wave demonstrated the work of a sample of a 28-qubit computer (Leda chip) online. Now the company is testing a prototype of a 128-qubit chip, which was code-named Rainier. The future plans of the company - to create in the near future a 1024-qubit computer.
All this, you see, far surpasses most other developments of quantum computers, and D-Wave was able to create a computer using semiconductor manufacturing technologies and existing semiconductor factories without resorting to optical circuits, quantum dots, laser containment or other exotic manufacturing technologies. D-Wave is also working on the second half of the problem, namely, on programming tools for creating applications that can take advantage of the possibilities that quantum computing promises to give.
The material in the D-Wave quantum chip is niobium; cool it to a sufficiently low temperature, and it will become a superconductor. When ordinary metal conducts an electric current, electrons, carriers of electric charge, collide with a non-ideal metal structure, resulting in resistance. When you cool a superconducting metal like niobium, the metal electrons form Cooper pairs. When Cooper pairs enter Josephson junctions on a chip (consisting of two superconducting niobium segments connected by a weak insulating barrier), they can be thought of as electron-like quasi-particles that can tunnel through an insulator in a junction, effectively conducting a current through it.
Niobium is located in the form of rings through which current can flow clockwise, against it or mixed, in both directions - corresponding, according to Rose, “0”, “1” or in a superposition of two values in the quantum bit of information (qubit), on which quantum computations are based. “The chip is a sequence of metal tracks on a silicon substrate; the substrate here is the same as that used for any semiconductor process, but on top are layers of metal separated by an insulator. Before us is a completely metallic magnetic device, where all information is stored as directions of current flow through metal loops and transitions. ”
The current direction is converted to the qubit value, depending on whether the qubit has an offset towards one direction (0 or 1), whether neighboring qubits move in the same or opposite direction, as well as from the energy barrier between different qubit states. The current Leda chip is equipped with 28 rings, which gives 28 qubits, but they are not connected with each other, only with a certain number of "neighbors". The Cooper pair in niobium is, precisely speaking, bosons, so they all exist in the same quantum state, Rose explained, which gives the entire superconductor quantum properties even without combining each qubit. Reducing the number of interconnects simplifies production.
In the experiments, the accuracy of quantum computing was 99.98%. Today it is the best result in the world. As you can see, our country also has something to brag about;)
For which problems, quantum computations give a gain compared to classical ones?
First , a quantum system effectively “solves” a complex computational problem — it models itself. On a quantum computer, you can model any quantum system in a polynomial number of steps. This will allow (in the presence of a quantum computer) to predict the properties of molecules and crystals, to design microscopic electronic devices with a size of several tens of angstroms. Now such devices are at the limit of technological capabilities, but in the future they are likely to be used in conventional computers.
The second example is factoring and similar number-theoretic problems associated with Abelian groups. About Shor’s algorithm has already been said before - the supposed complexity of the factorization task lies at the heart of the cryptographic strength of some public key encryption algorithms, such as RSA. And this beautiful result can have more harmful than useful use: decomposing numbers into factors, you can pick up keys to ciphers, forge electronic signatures, etc. However, the difficulties in creating a quantum computer are so great that over the next 10 years, cipher users can sleep peacefully.
The third example is the search for the desired record in an unordered database (Grover’s algorithm) and similar tasks (for example, image recognition — something that D-Wave demonstrated on its first chips). In the future, a quantum computer can bring us closer to solving the problem of creating artificial intelligence.
Quantum computers are the "holy grail" of modern physics. The idea of a quantum computer looks as tempting as it is unrealistic. Probably also perceived the project of a regular computer in the time of Charles Babbage, the invention of which was implemented only a hundred years later. The problem of creating a quantum computer is equivalent in complexity to the problem of interstellar flights and the problem of thermonuclear fusion. QC on two or three qubits already exist now, but they also require high technologies for their construction (very pure substances, very precise implantation of individual atoms, a highly accurate measurement system). But, as mentioned earlier, the main challenge is not technological, but the fundamental one is scalability.
Hopefully, in our time, scientific and technical progress is going faster, so do not have to wait that long. Perhaps one fresh idea plus a few years to develop a new technology is enough ...
1) Mensky M.B. - Man and the quantum world. Moscow, Vek2, 2007. ISBN: 5-85099-161-1
2) Google Tech Talks series on Quantum Computing Neart and Geordie Rose at Google | Santa Monica, CA
3) Wikipedia articles
4) Computerry Articles
5) S.YA. Killeen. Quantum information. UFN, t. 169, No. 5, p. 507-527, 1999
6) Quantum educational program. Lenta.ru
7) Tom's Hardware guide. D-Wave Orion: the first quantum computer. Dmitry Chekanov, August 6, 2008
8) 8) A.Kh. Shen, M.N. Sluggish. Course of lectures "Classical and quantum calculations". Internet University of Information Technology
9) Quantum computer and quantum computing. Ed. Sadovnichy V.A.
10) Quantum Computation: Pros and Cons. Ed. Sadovnichy V.A.
A bit of physics
To be consistent, it is impossible not to mention the theoretical achievements of quantum physics, which made it possible in practice to create such devices.
In general, what is quantum mechanics? It is known that quantum mechanics is a theory by which microscopic systems obey, for example, atoms. The microworld - atoms, electrons, photons and other particles - lives according to special laws. There, not just everything is very small, everything is completely different there, and many phenomena of the microworld have no analogues in the usual macrocosm, which makes them seem fantastic.
A characteristic feature of quantum mechanics, which is imprinted in the name - in some physical systems, energy is quantized , that is , it can be equal to one of a certain discrete set of numbers, or even any other value (forbidden zone). If you try to imagine such a situation in the macrocosm, you can imagine, for example, that objects have a temperature that is expressed only by an integer number of degrees. That is 10, 20, 31, 36 degrees - maybe, but 36.6 - is simply impossible. It is possible to heat and cool items, but at the same time the temperature will jump back and forth immediately to a degree.
')
In particular, the energy of the electromagnetic field is emitted only in the form of discrete indivisible portions. Such a portion is called a quantum . Max Planck made the assumption of the existence of quanta in 1900–1901, thereby laying the foundation for quantum theory, quantum mechanics, and much more with quantum adjectives - including computers.
Another amazing property of photons, electrons and other particles is that they can exhibit the properties of both particles and waves . Light presented in classical physics as an electromagnetic wave, turned out in quantum physics as a set of photons, like particles. But conversely, particles, such as electrons, exhibit properties characteristic of a wave: they interfere with each other and diffract on obstacles (that is, bend around them). This property of particles gave the basis in the early years to call the theory describing them wave mechanics, and only later the name "quantum mechanics" was established.
And finally, the third fundamental property: a quantum object, in contrast to the classical one, is initially statistical . It is possible to describe the location of a particle not directly observable in a certain place only with a certain probability.
There is also a significant limitation on observations and measurements in the microworld: the Heisenberg uncertainty principle : it is impossible to measure simultaneously the speed and coordinates of particles.
As long as we do not measure objects, they behave even worse. If a quantum system can be in several states and it is not known in which it is located, then we are talking about a superposition of states . It can be said that it is unknown what state the system is in, or that it is in several states at the same time, it is a matter of interpretation. In any case, when measuring, the system selects one of the states.
A well-known illustrative example is a thought experiment called “Schrödinger's cat”: a live cat, a container with a poisonous gas and a radioactive core are placed in a closed box. If the nucleus disintegrates, it activates a mechanism that opens the gas tank and thereby kills the cat. The probability that the core will disintegrate in an hour is 50 percent. An hour later, the cat in the box is alive with a probability of 50 percent. From the point of view of quantum mechanics, while the box is closed, the cat is in a superposition of two states (either alive or dead; and alive, and dead; neither alive nor dead - however you please). At that moment, when the observer opens the box, he sees whether the cat is alive or dead.
Finally, another important phenomenon for us is quantum entanglement, which is also confusion, cohesion, and sometimes coherence. They say about entanglement when the state of two (or more) quantum systems must be described in conjunction with each other, even if the systems themselves are spaced apart. Accordingly, the physical properties of each of the systems are associated with the physical properties of the other, while they may not be close by and not be connected in any way.
If two entangled systems are in a superposition of states, then by measuring the state of one, you can find out the state of the other . For example, you can confuse two atoms, the spin (a certain quantum characteristic) of one of which will be directed up, and the other down, and we will not know which atom has which spin. But by measuring the spin of one atom, we immediately recognize the spin of another, even if they are spaced apart.
All these and many other features of the microworld make it possible to build a quantum computer.
Classic and quantum bit information
A quantum computer, like any other computing device, must operate with numbers. The simplest device capable of representing numbers is a device that can be in two stable states. So, conductors with a current can be in two stable states: when there is no current, it corresponds to a value of 0, and when there is a current, it corresponds to a value of 1. There should be no other steady state. The use of such devices for conducting operations on numbers is possible due to the binary recording of numbers — information such as the choice between two options (digits 0 and 1 or, equivalently, two system states) is stored in such a cell. It is said that the amount of information stored in a binary cell is 1 bit. Two binary cells store 2 bits of information. This corresponds to the choice between 2-bit binary numbers 00, 01, 10 and 11. In a register consisting of N binary cells, N bits of information are stored, which corresponds to the choice of one N-digit binary number of all possible (all such numbers are 2 ^ N ).
Now let's see how the storage of information in a quantum computer is different. The information itself can also be represented in the form of binary numbers, that is, chains of zeros and ones. One digit of a binary number is stored in a binary cell. This cell is called a qubit and is a quantum system, one of the states of which corresponds to the digit 0, and the second - to the digit 1. We denote these states as ψ1 and 2. There are still no differences from the classic computer. But it is at this point that a significant difference appears. The fact is that for quantum systems the so-called superposition principle holds: if a quantum system can be in one of two states, say, in the state ψ1 and in the state ψ2 , then it can also be in the whole family of states that are constructed as linear combinations of the original. This means that the initial states are multiplied by some (complex) numbers and added together. If the numbers c1 and c2 are chosen, then the state is c1ψ1 + c2ψ2 .
In particular, if a qubit is brought into the state (∣0〉 + ∣1〉) / √2 , then this means that both the digit 0 and the digit 1 are simultaneously written in it.
Quantum parallelism
So, in one binary cell of a quantum computer, called a qubit, can be stored not only one of the two digits of the binary number, 0 or 1 (as it would be in the case of a classical computer), but both of these numbers at the same time. For example, 4 binary numbers 00, 01, 10 and 11 can be stored in two qubits simultaneously. And if some register of a quantum computer contains N qubits, then such a register can simultaneously store 2 ^ N binary numbers of length N. And when a quantum computer acts all these numbers are processed simultaneously. This is quantum parallelism. If we had only classical computers at our disposal, each of which works with binary numbers of length N, then 2 ^ N computers would be necessary for simultaneous processing of 2 ^ N such numbers. If we managed to build a quantum computer containing N qubits, then one (!!!) this computer simultaneously processes all 2 ^ N numbers.
Quantum algorithms
Unlike classical computers , quantum computers are not universal : there is no algorithm for solving computational problems on a quantum computer for any computing task. So far, only a small number of quantum algorithms have been found.
The most famous quantum algorithms:
1) Shor Algorithm (factorization)
2) Grover's algorithm (quick search in an unordered database)
3) Algorithm Deutsch - Jos (answer to the question, constant or balanced function)
Due to the use of the quantum entanglement phenomenon and the principle of superposition, quantum algorithms have a significant increase in execution speed compared to the corresponding classical algorithms.
Shor Algorithm - Killer public-key cryptography
In a bit more detail, let's stop on Shor's algorithm, which allows to solve any of two mathematically equivalent problems:
1) find the period of a complex periodic function or
2) decompose a very large number into prime factors
The second of these tasks is of significant practical importance because it is used in cryptography. The fact is that in one of the cryptographic methods for encrypting and decrypting secret messages (public key encryption) they use large numbers for which their factorization is known. It is clear that such numbers are easy to obtain: it is enough to multiply a large number of prime numbers, and we get a very large number for which the factorization is known. The recipient of the encoded secret message can decode it because the decoding procedure uses factorization of a long number, and he knows this decomposition.
If the enemy could factor this number into prime factors, he would also be able to decode the message. However, this decomposition takes a lot of time. Therefore, from a practical point of view, it is impossible to decode such a message. But if the enemy had a quantum computer (of sufficient power, that is, working with sufficiently large numbers), then he could decompose long numbers into simple factors and therefore could easily decipher such messages. This common cryptography method would stop working. This is one of the arguments that make the creation of a quantum computer important.
Problems of building a quantum computer
Physical:
- decoherence (the main problem today, it will be discussed later)
- search for specific processes performing logical operations
Mathematical:
- search for new algorithms
- search for error correction methods
These problems can be described in one word - scalability . Attaching additional memory to a regular computer is a simple routine procedure. Attaching each new qubit to QC is for now a piece work.
Decoherence
The phenomenon of decoherence is that the system being measured loses its specific quantum properties. In other words, the pure state quickly turns into a mixture when the quantum system interacts with the environment.
To combat decoherence, various methods of isolating a quantum system are developed, on the one hand, including the use of extremely low temperatures and high vacuum, and on the other, an introduction to quantum computing of codes that are resistant to decoherence errors (usually in such circuits encoded by the state of several connected physical qubits).
QC requirements
The accuracy of unitary operators, according to various estimates, should be from 99.99 to 99.999999 (depending on the specific implementation).
Modern options for implementing a quantum register
- ion traps
- superconductors
- quantum dots
- NMR
- photons
- many other exotic ideas (anions, etc.)
Ionic traps are the first and most well-developed idea. One of the superconducting computers will be discussed further. Well, quantum dots are considered the most promising option.
D-Wave Superconducting Quantum Computers
In February 2007, the Canadian company D-Wave Systems assembled a 16-qubit quantum computer, which founder and CEO of Georgy Rose called the most powerful quantum computer ever built, and the first one that can run commercially-significant applications. In November 2007, at the conference dedicated to supercomputers, the same company D-Wave demonstrated the work of a sample of a 28-qubit computer (Leda chip) online. Now the company is testing a prototype of a 128-qubit chip, which was code-named Rainier. The future plans of the company - to create in the near future a 1024-qubit computer.
All this, you see, far surpasses most other developments of quantum computers, and D-Wave was able to create a computer using semiconductor manufacturing technologies and existing semiconductor factories without resorting to optical circuits, quantum dots, laser containment or other exotic manufacturing technologies. D-Wave is also working on the second half of the problem, namely, on programming tools for creating applications that can take advantage of the possibilities that quantum computing promises to give.
The material in the D-Wave quantum chip is niobium; cool it to a sufficiently low temperature, and it will become a superconductor. When ordinary metal conducts an electric current, electrons, carriers of electric charge, collide with a non-ideal metal structure, resulting in resistance. When you cool a superconducting metal like niobium, the metal electrons form Cooper pairs. When Cooper pairs enter Josephson junctions on a chip (consisting of two superconducting niobium segments connected by a weak insulating barrier), they can be thought of as electron-like quasi-particles that can tunnel through an insulator in a junction, effectively conducting a current through it.
Niobium is located in the form of rings through which current can flow clockwise, against it or mixed, in both directions - corresponding, according to Rose, “0”, “1” or in a superposition of two values in the quantum bit of information (qubit), on which quantum computations are based. “The chip is a sequence of metal tracks on a silicon substrate; the substrate here is the same as that used for any semiconductor process, but on top are layers of metal separated by an insulator. Before us is a completely metallic magnetic device, where all information is stored as directions of current flow through metal loops and transitions. ”
The current direction is converted to the qubit value, depending on whether the qubit has an offset towards one direction (0 or 1), whether neighboring qubits move in the same or opposite direction, as well as from the energy barrier between different qubit states. The current Leda chip is equipped with 28 rings, which gives 28 qubits, but they are not connected with each other, only with a certain number of "neighbors". The Cooper pair in niobium is, precisely speaking, bosons, so they all exist in the same quantum state, Rose explained, which gives the entire superconductor quantum properties even without combining each qubit. Reducing the number of interconnects simplifies production.
Russian technologies: laboratory of quantum computers of FTIAN (Physical-Technical Institute of RAS)
- quits based on nuclear spin chains. Using modern technology of creating structures in a semiconductor with a size of several nanometers, we can implant a linear chain of phosphorus ions into a narrow channel in silicon. One such chain contains from ten thousand to a million atoms (the nuclear spin is very small, and in order to control it and reliably detect it, you need to collect a sufficiently large number of particles). This is one logical qubit.
- high-temperature superconductors (HTSC). p-contact. In this case, a qubit is created on the border of two different types of HTS (SDS transition). The potential energy of such a transition has two minima, with the values of the phase variable 0 and p. They correspond to two basic qubit states.
- ion trapped qubits
- optical-controlled quantum dots
In the experiments, the accuracy of quantum computing was 99.98%. Today it is the best result in the world. As you can see, our country also has something to brag about;)
Why do you need a quantum computer?
For which problems, quantum computations give a gain compared to classical ones?
First , a quantum system effectively “solves” a complex computational problem — it models itself. On a quantum computer, you can model any quantum system in a polynomial number of steps. This will allow (in the presence of a quantum computer) to predict the properties of molecules and crystals, to design microscopic electronic devices with a size of several tens of angstroms. Now such devices are at the limit of technological capabilities, but in the future they are likely to be used in conventional computers.
The second example is factoring and similar number-theoretic problems associated with Abelian groups. About Shor’s algorithm has already been said before - the supposed complexity of the factorization task lies at the heart of the cryptographic strength of some public key encryption algorithms, such as RSA. And this beautiful result can have more harmful than useful use: decomposing numbers into factors, you can pick up keys to ciphers, forge electronic signatures, etc. However, the difficulties in creating a quantum computer are so great that over the next 10 years, cipher users can sleep peacefully.
The third example is the search for the desired record in an unordered database (Grover’s algorithm) and similar tasks (for example, image recognition — something that D-Wave demonstrated on its first chips). In the future, a quantum computer can bring us closer to solving the problem of creating artificial intelligence.
Quantum computers are the "holy grail" of modern physics. The idea of a quantum computer looks as tempting as it is unrealistic. Probably also perceived the project of a regular computer in the time of Charles Babbage, the invention of which was implemented only a hundred years later. The problem of creating a quantum computer is equivalent in complexity to the problem of interstellar flights and the problem of thermonuclear fusion. QC on two or three qubits already exist now, but they also require high technologies for their construction (very pure substances, very precise implantation of individual atoms, a highly accurate measurement system). But, as mentioned earlier, the main challenge is not technological, but the fundamental one is scalability.
Hopefully, in our time, scientific and technical progress is going faster, so do not have to wait that long. Perhaps one fresh idea plus a few years to develop a new technology is enough ...
Literature
1) Mensky M.B. - Man and the quantum world. Moscow, Vek2, 2007. ISBN: 5-85099-161-1
2) Google Tech Talks series on Quantum Computing Neart and Geordie Rose at Google | Santa Monica, CA
- Introduction to Quantum Computing
- Image Recognition with an Adiabatic Quantum Computer
- Does it need to refer to quantum mechanics?
3) Wikipedia articles
4) Computerry Articles
- Quanta for quanta. Computerra №23 dated June 21, 2006
- D-Wave: qubit show with morality. Computerra №9 dated March 9, 2007
- The space of freedom (interview Yuri Manin) "Computerra" №2 dated January 22, 2001
5) S.YA. Killeen. Quantum information. UFN, t. 169, No. 5, p. 507-527, 1999
6) Quantum educational program. Lenta.ru
7) Tom's Hardware guide. D-Wave Orion: the first quantum computer. Dmitry Chekanov, August 6, 2008
8) 8) A.Kh. Shen, M.N. Sluggish. Course of lectures "Classical and quantum calculations". Internet University of Information Technology
9) Quantum computer and quantum computing. Ed. Sadovnichy V.A.
10) Quantum Computation: Pros and Cons. Ed. Sadovnichy V.A.
Source: https://habr.com/ru/post/95428/
All Articles