Unusually connecting the qubits to each other, D-Wave significantly increased the speed of the quantum computer.

“The Great Wave in Kanagawa” - a 19th century Japanese woodcut by Katsushiki Hokusai

In early March, D-Wave Systems announced the release of their new computer, operating on the principle of quantum annealing . In the new machine made several technical improvements, as well as significantly changed the physical location of the components. What does this mean? Together with the online resources of the company D-Wave, the device, approaching the state of usefulness, begins to take shape.

Make a sleek computer

Before you get to the delicious filling, you first need to chew on the edges of the cookies - that is, find out, what is quantum annealing? Most computers work in a straightforward way: to add two numbers, we create a set of logic gates that perform addition. Each of the gates performs a set of its own, well-defined operations on the input data.

But this is not the only way to do calculations. Most of the tasks can be written so that they are equivalent to the problem of energy minimization. In this embodiment, the task is an energy landscape, and the solution is the minimum possible energy on it. The point is to find a combination of bit values ​​denoting this energy.
')
To do this, start with a flat energy landscape: all bits will have minimal energy. Then we slowly and carefully change the terrain around the bits until it begins to represent our task. If done correctly, the bits will remain in a state with minimal energy. The solution we get by counting their values.

Although it all works without quantum physics, D-Wave does it with quantum bits (qubits). This means that qubits correlate with each other — this is called quantum entanglement. As a result, they change their values ​​together, not separately.

Tunneling

As a result, an effect known as quantum tunneling becomes possible. Imagine that a qubit is stuck in a state of high energy. Nearby there is a state with less energy, in which I would like to move the qubit. But to get there, he first needs to go into a state with more energy. In the classical system, this turns into a barrier on the way to achieving a state with less energy. But in a quantum qubit it can tunnel through the energy barrier, having come to a state with less energy.

Two of these properties may allow such a computer, which is controlled by D-Wave, to find solutions to some problems faster than the classic one.

But the devil is in the details. In the computer, the energy landscape is constructed by the binding (physical union) of qubits. Binding controls how much the value of one qubit affects the value of the rest.

This moment has always been a problem for the car from D-Wave. Under ideal conditions, every qubit will have connections with every other qubit. But to organize such a large number of connections is impractical.

Kubbit by itself

The consequences of a lack of connections are very serious. Some tasks simply cannot be altered to solve on D-Wave machines. And sometimes, in cases where the task can be remade, the calculations will be inefficient. Imagine that solving a problem requires connecting qubits with numbers one and three, but they are not directly connected. In this case, you have to look for qubits common to both of them. Suppose a qubit one is connected to a qubit five, and a qubit two is connected to qubit five and three. Then the logical qubit one will be the combination of the first and the fifth. Logical qubit three - a combination of the second and third. D-Wave calls this sequence the length of the chain. In this case, the length is two.

Due to the linking of physical qubits to logical qubits, fewer qubits remain available for computation.

D-Wave was planning to build even more complex qubit schemes to increase connectivity. The greater the connectivity, the shorter the length of the chains, the more free logical qubits. And if qubits are tightly connected together, and the connectivity is great, then using such a machine you can solve more problems.

The efficiency of structuring some tasks will be extremely low, that is, the D-Wave architecture is simply not adapted to solve them. But with increasing connectivity, the number of inappropriate tasks will decrease.

In the previous version of the machine, the qubits were distributed in blocks of eight pieces in order to improve the connectivity of the diagonal blocks compared to the previous version of the machine. As a result, the situation with the length of the chains has improved somewhat.


D-Wave 2000Q Architecture

D-Wave has now switched to a connectivity scheme known as the “Pegasus graph”. I do not know how to describe it accurately, so I will describe it not very correctly from the point of view of strict graph theory, but more clearly. Instead of the same blocks of eight qubits in the machine, there are now two types of blocks: eight pieces and two pieces each.

In blocks of eight, the qubits are located, as before, along the inner and outer loops. But, as shown in the video, now the inner and outer loops have additional links. This means that each qubit in a small block has five links.

The blocks themselves are no longer lined up in the right grid, and there are more connections between qubits from different blocks. In the previous generation, the qubits on the outer loops were connected with other qubits on the outer loops, and now each qubit is associated with both internal and external loops of neighboring blocks.



In addition, a new network of long-distance communications between different blocks. Each qubit has a relatively far connection with another qubit in the remote block. The density of long-range connections increases due to the second main building block, consisting of a combined pair of qubits. Couples are located around the main units and complement long-distance connectivity.

The idea is that united in groups of eight qubits, located on the edge of the chip, the density of bonds is almost the same as that of the internal groups, unlike graphs of the “chimera” class.

Shortening chains

What does all of this mean? First, the similarity of the “chimera” and “pegasus” graphs means that the code developed for the “chimera” should also work on “pegasus”. Increasing connectivity means reducing the length of chains and increasing the reliability of calculations.

So that you can imagine how much the new graph improves the situation, I will say that a square lattice with diagonal connections requires chains of six units in graphs of the “chimera” type, and two units in graphs of the “pegas” type. In general, the length of the chains is reduced by two or more times. As a result, the work time is reduced by 30-75%.

In addition to the new graph, D-Wave has improved the performance of the computer at the technical level: the noise level of qubits is less, and their number has increased significantly. The company plans using the new architecture to bring the number of qubits to 5,000 (from 2000). All these architectural changes mean that much more physical qubits can be used as independent logical qubits, so the upgrade will be much more significant.