MISIS (prototype of the quantum computer)

2019: Prototype of the quantum computer

On October 2, 2019 it became known that in NITU "MISIS" the Russia's first prototype of the quantum computer earned. The device on two qubits executed the set algorithm, having exceeded earlier known limit of accuracy for 3%. As a basis for qubits superconducting materials were taken.

Works on creation of the quantum computer within the project of Fund of perspective researches are conducted in NITU "MISIS" since 2016 under the leadership of Valery Ryazanov, the chief researcher of Laboratory of superconducting metamaterials of the university. Construction assumes use as a basis for qubits of superconducting materials.

Qubits (quantum bits) – the operating force of the quantum computer, an analog of "bits" at the normal PC, only much more perfect. If usual to us the computer "thinks" and considers zero and units, i.e. each bit of information can be coded either as "0" or as "1", then the qubit has property of so-called superposition, capability to be at the same time in both statuses. It offers huge prospects, at such computing resources the quantum computer will be able to overtake the most powerful computing devices on the whole orders.

The quantum computer on superconducting materials – more perfect system in comparison with analogs. For example, other research teams develop qubits on separate atoms (which can "be lost" because of it is insignificant the small size) and on ions (it is possible to build them exclusively linearly that physically inconveniently). The qubits created in NITU "MISIS" are made of aluminum, have the size in 300 microns, they cannot be "lost", and still it is possible to build not linearly.

During the experiment the dvukhkubitny quantum computer solved Grover's algorithm – a search algorithm for function. The quantum computer, thanks to the superposition principle, ideally can find the correct value x in the solution of this task for one appeal to the f (x) function with the probability of 100%.

Grover's algorithm on two qubits is very important step on the way to creation of the quantum computer. We showed its work, but there is a speech first of all about technology achievement. We showed sales opportunity of all necessary for logical actions for the universal quantum processor: initialization, odnokubitny and dvukhkubitny transactions and reading, and with the level of errors, satisfactory for small algorithms, told one of project participants engineer of Superconducting Metamaterials laboratory Ilya Besedin

The most great difficulty on the way to creation of the useful quantum processor – errors. Unlike klasschesky computers which can work for years and always issue the reproduced and predictable results, quantum computers are subject to influence of noise which distorts results of calculations. In spite of the fact that the processor created in NITU "MISIS" from two qubits is too small for the solution of applied tasks, it successfully "stepped" over a threshold of 50% probability of the right answer, having reached 53%.

All algorithm consists of initialization of two qubits, four odnokubitny transactions, two dvukhkubitny transactions and reading of two qubits; errors in any of which are reduced by the probability of the correct answer in the answer.

The chip for the quantum computer was manufactured in MSTU of Bauman, and were already engaged in his design and start of the device in NITU "MISIS" where in Superconducting Metamaterials laboratory the equipment with the cryostats ensuring functioning at ultralow temperatures to-273.14 Celsius degrees that is near absolute to zero is built.

Nevertheless, before us still a big way. Quite recently got to the press yet article of Google company which is not published officially which managed to implement an algorithm of "quantum superiority" on the 53-kubitny superconductor quantum processor. The problem of "quantum superiority" is a task, optimum for the quantum computer, which at the same time is very difficult for executing on the classical computer. And if we have an overcoming a "classical" limit - it is after all fundamental result, then result of Google – it is already closer in a practical side: they could formulate and solve a problem which their processor can execute in minutes, and the powerful supercomputer checked for weeks, adds Ilya Besedin

And even at this Google it was not succeeded to approach that the quantum computer solved any practically useful problem more effectively, than classical. However still theoretical predictions of rather computing superiority of quantum computers ekspriment are confirmed.

The following important steps on the way to creation of the useful quantum computer are a demonstration of the "useful" quantum algorithms reduced to the sizes of several tens qubits of the version (for example, the simulator of chemical reaction or the main status of a molecule) and demonstration of quantum error correction. That's it, by the way, superconductor qubits are suitable for error correction better in total them it is possible to organize in a two-dimensional grid with local interactions and parallel valves which is necessary for "the superficial code" - the simplest in terms of requirements and to the accuracy of transactions.

We want to move in this direction too, but from my point of view in quantum computings important not only it is "more", but also it "is better": superconductor qubits which we use turn out happy expensive and give many errors. And before doing hundreds and thousands of qubits, in my opinion, it is worth working over the most basic unit – qubit, sums up the result Ilya Besedin