Computing device

Abstract

To provide a computing device that eliminates a communication bottleneck in distributed quantum computation.SOLUTION: A computing device includes a first ring node where multiple nodes are connected in a ring shape, and a second ring node where multiple nodes are connected in the ring shape, Communication is performed between a node in the first ring node and a node in the second ring node by consuming quantum entanglement between the node in the first ring node and the node in the second ring node.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to distributed quantum computing. [Background technology]

[0002] Quantum computers are a technology that performs calculations by utilizing the principle of superposition in quantum mechanics, and are expected to be able to quickly solve problems such as prime factorization and quantum chemistry calculations. For this reason, their development is being actively pursued around the world. A (classical) bit, which is an element that makes up a classical computer, takes on a value of 0 or 1. On the other hand, a qubit, which is an element that makes up a quantum computer, can take on a continuous superposition state of 0 and 1 in addition to 0 and 1. Using this superposition state, it is possible to simultaneously perform calculations where the bit value is 0 and 1, but when the qubit is observed, its value is fixed at 0 or 1, and the superposition state is destroyed.

[0003] While various elements have been proposed as devices for constructing quantum computers, the size that can be realized on a single node is limited. In particular, when quantum error correction, which uses multiple qubits to construct error-tolerant logical qubits, is incorporated, the number of error-tolerant qubits that can be implemented on a single node may be limited to just a few. To achieve large-scale quantum computing beyond the limitations of a single node, distributed quantum computing, in which multiple quantum computer nodes are connected via quantum communication channels and calculations are performed while communicating, is effective. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Rodney Van Meter, Simon J. Devitt, "Local and Distributed Quantum Computation," May 2016. https: / / arxiv.org / abs / 1605.06951 [Non-patent document 2] Rodney Van Meter, Kae Nemoto, WJ Munro, "Communication Links for Distributed Quantum Computation," Jan 2007. https: / / arxiv.org / abs / quant-ph / 0701043 [Non-patent document 3] Anbang Wu, Hezi Zhang, Gushu Li, Alireza Shabani, Yuan Xie, Yufei Ding, "AutoComm: A Framework for Enabling Efficient Communication in Distributed Quantum Programs," Oct 2022. https: / / arxiv.org / abs / 2207.11674 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in conventional distributed quantum computing, if communication between specific nodes becomes a bottleneck, the overall computation speed of the distributed quantum computing is limited by the communication bandwidth, meaning that performance cannot be improved even by using multiple nodes.

[0006] The present invention has been made in view of the above points, and aims to provide a technique for eliminating communication bottlenecks in distributed quantum computing. [Means for solving the problem]

[0007] According to the disclosed technology, a first ring node in which a plurality of nodes are connected in a ring shape; a second ring node in which a plurality of nodes are connected in a ring shape; Communicating between a node in the first ring node and a node in the second ring node by consuming quantum entanglement between the node in the first ring node and the node in the second ring node. A computing device is provided. [Effects of the Invention]

[0008] The disclosed technology provides a technology for eliminating communication bottlenecks in distributed quantum computing. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 illustrates three protocols. [Figure 2] FIG. 1 is a diagram illustrating an example of a basic device configuration. [Figure 3] FIG. 2 is a diagram illustrating an example of a node configuration. [Figure 4] FIG. 10 is a diagram illustrating a configuration in which communication is performed between single nodes. [Figure 5] FIG. 2 illustrates a ring node. [Figure 6] FIG. 10 is a diagram illustrating a configuration when communication is performed between ring nodes. [Figure 7] FIG. 1 is a diagram for explaining a processing procedure in the first embodiment. [Figure 8] FIG. 1 is a diagram for explaining a processing procedure in the first embodiment. [Figure 9] FIG. 10 is a diagram illustrating a quantum network (computing device) in a second embodiment. [Figure 10] FIG. 10 is a diagram illustrating a specific example of a process for determining a path. [Figure 11] 10 is a flowchart of processing by a computing device. [Figure 12] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 13] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 14] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 15] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 16] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 17]FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 18] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 19] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 20] FIG. 10 is a diagram for explaining a processing procedure in the second embodiment. [Figure 21] FIG. 2 illustrates an example of a hardware configuration of the apparatus. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention (the present embodiment) will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

[0011] Below, the problem will first be explained in more detail, and then the configuration and operation of the device according to this embodiment will be explained in detail.

[0012] (About the assignment) A new problem that arises when implementing distributed quantum computing is the issue of communication between nodes. To exchange quantum bit information between nodes, quantum communication between nodes is required. In this respect, quantum distributed computing differs significantly from regular distributed computing in two respects.

[0013] The first difference is that quantum states cannot be copied due to the no-cloning theorem, so load balancing techniques that rely on copying information and distributing it to multiple nodes cannot be used in principle.

[0014] The second difference is that the bandwidth used to send quantum information can be pooled as a resource. Behind this, three protocols are important: quantum entanglement generation, entanglement swapping, and quantum teleportation. Figure 1 shows an image of these three protocols. The figures in Figure 1 represent nodes.

[0015] Quantum entanglement generation, shown in Figure 1(a), is a procedure in which two distant nodes communicate with each other and share a quantum-correlated two-qubit state (e.g., a Bell state) called a quantum entanglement state between the two nodes. Quantum entanglement generation is achieved by creating correlated two-qubit information on one node and transmitting one of the two qubits to the other node. However, since such quantum information transmission generally involves a large amount of noise, generating quantum entanglement between nodes requires transmitting the information multiple times and using a protocol called quantum entanglement distillation to generate a quantum entanglement state with low noise.

[0016] Quantum teleportation in Figure 1(b) is a protocol for transferring quantum bits from one node to another by consuming quantum entanglement placed at two nodes and performing classical communication.

[0017] Entanglement swapping in Figure 1(c) is a protocol that generates quantum entanglement between A1 and An by consuming the entanglements on this path one by one when quantum entanglement is generated in a chain from node A1 to A2, A2 to A3, ..., An-1 to An.

[0018] Therefore, if you want to transfer quantum data between node A1 and node An, you can generate quantum entanglement along the route from A1 to A2, A2 to A3, and so on, from An-1 to An, consume these quantum entanglements to establish quantum entanglement between A1 and An, and then consume the quantum entanglement between A1 and An to teleport the quantum bit of A1 to An.

[0019] In this case, the time required for entanglement swapping and quantum teleportation can generally be considered negligibly short compared to the time required for quantum entanglement generation. Furthermore, among the above protocols, the generation of quantum entanglement can begin before the information to be sent and the route are decided. Therefore, in quantum communication, the procedures required to send quantum bits can be effectively executed ahead of time, even before the information and route of the quantum bits to be sent are decided. This is a feature not found in ordinary communication or distributed computing.

[0020] As mentioned above, quantum entanglement generation is expected to be extremely slow compared to normal communication and calculations within quantum computing nodes, and when communication procedures are performed continuously between two nodes, the speed is limited by the speed of quantum entanglement generation. If it is possible to connect any number of communication paths depending on the procedures required between each node, the bandwidth can be improved by preparing multiple communication paths for generating quantum entanglement.

[0021] However, in terms of implementation, it is technically difficult to integrate the communication channels used for quantum entanglement generation, and it is expected that only a small number of communication channels can be placed at each node in reality. If the speed at which quantum entanglement can be generated per channel and the number of communication channels that can be connected per node are constrained by certain constants, then naively speaking, the bandwidth between nodes also becomes constant. If the communication demands exceed this, communication becomes a bottleneck and even distributed processing cannot increase the computation speed. For this reason, distributed quantum computing is considered useful for expanding the systems it implements, but when communication becomes a bottleneck, conventional distributed computing methods cannot solve this problem, and it has the drawback of incurring a large overhead in speed.

[0022] In distributed quantum computing, when each node can only have communication paths with a fixed number of nodes (three or more), and the bandwidth of each communication path is fixed, the amount of communication between two specific nodes is at most a fixed amount. If this problem causes communication between specific nodes to become a bottleneck, the overall computation speed of the distributed quantum computing will be limited by the communication bandwidth, and performance cannot be improved even by distribution.

[0023] (Outline of the embodiment) In this embodiment, a network with effectively high throughput is constructed by regarding the transmitting and receiving nodes as virtual nodes multiplexed in a ring. This makes it possible to improve the bandwidth between two nodes by increasing the number of nodes, and to eliminate the bottleneck of distributed quantum computing while trading off the number of nodes.

[0024] The configuration and operation of the device according to this embodiment will be described in detail below.

[0025] (Device configuration) FIG. 2 shows a basic device configuration in this embodiment. As shown in FIG. 2, in this embodiment, a quantum network 1 consisting of a plurality of nodes performs quantum computation. This quantum network 1 may be called a "computing device." Also, as shown in FIG. 2, a control device 2 that controls the quantum computation in the quantum network 1 may be provided outside the quantum network. The "quantum network 1 and control device 2" may be called a "computing device." The "quantum network 1 and control device 2" may be called a "computing system."

[0026] 3 shows an example of the configuration of a node 100 that constitutes the quantum network 1. The node 100 may also be called a quantum computer. As shown in FIG. 3, the node 100 includes a control device 110 and a quantum processor 110.

[0027] The quantum processor 110 includes a physical system that realizes quantum bits that hold data, a function for generating quantum entanglement, a function for performing measurements and operations to realize quantum teleportation, and the like.

[0028] The control device 110 instructs the quantum processor 120 to perform measurements and operations, acquires measurement results, performs calculations based on the measurement results, notifies other nodes of the measurement results, etc. The control device 110 may be a classical computer.

[0029] Each node (and the quantum network 1 (=computing device)) is not limited to a device that uses quantum bits of an actual physical system. Each node (and the quantum network 1 (=computing device)) may be a virtual one simulated on a classical computer.

[0030] The control for executing the protocol of the first or second embodiment described below may be performed by the control device 110 of any one of the nodes, or may be performed in cooperation by a plurality of control devices 110 of a plurality of nodes.

[0031] Furthermore, the control device 2 shown in FIG. 2 may execute the protocol of the first or second embodiment by transmitting instructions to each node that configures the quantum network 1.

[0032] The configuration and operation of a specific quantum network 1 (computing device) in this embodiment will be described below. In this embodiment, two examples will be described: a simple example 1 that shows an improvement in bandwidth at two nodes in distributed quantum computing with nodes having at most three communication paths, and an example 2 that can withstand more complex accesses in distributed quantum computing with nodes having at most four communication paths.

[0033] Example 1 First, a description will be given of Example 1. In Example 1, a description will be given of distributed quantum computation between two ring nodes of a node having three communication paths.

[0034] FIG. 4 shows communication between single nodes. In the first embodiment, as shown in FIG. 5, a string of n nodes arranged in a ring shape is considered as one node, which is called a ring node. The ring node in the example shown in FIG. 5 consists of four nodes. The elements that make up the ring node are called child nodes. In the first embodiment, communication is performed between ring nodes as shown in FIG. 6. FIG. 6 shows an example of the configuration of a computational device (quantum network) in the first embodiment.

[0035] In Example 1 (and Example 2), two ring nodes are named a and b, and the i-th child node of each ring node is named a. i ,b i In FIG. 6 (and the following figures), the child nodes are given the symbol i.

[0036] Here, to simplify the equation, n =a0,b n = b0, the communication channel is 0≦i <nについて、(a i ,a i+1 ), (b i ,b i+1 ), (a i ,b i ) are pasted between the nodes. In the initial state, the data used for the calculation is placed in the child nodes a0 and b0 of each node, and other nodes do not hold the data used for the calculation. In this case, by following the steps S1 (step 1) to S4 (step 4) below, it is possible to perform calculations without the communication bandwidth becoming a bottleneck.

[0037] S1: The calculation device sets the variable i=0.

[0038] S2: The computing device is a i and b i Among the data held in (a i ,b i ) to perform calculations by consuming the quantum entanglement.

[0039] S3: The computing device (ai ,a i+1 ) using quantum entanglement i The data of a i+1 The computing device also transfers the i ,b i+1 ) and use the quantum entanglement of b i The data of b i+1 Transfer to.

[0040] S4: The computing device updates the variable i to i+1 mod n and returns to S2.

[0041] A specific example of the procedure for communication when n=4 will be explained with reference to Figures 7 and 8. In step 1, data is placed in each of the child nodes a0 and b0. In step 2, an operation is performed between the data held in each of the child nodes a0 and b0, consuming the quantum entanglement of (a0, b0).

[0042] In step 3, the data of a0 is transferred to a1 using the quantum entanglement of (a0, a1), and the data of b0 is transferred to b1 using the quantum entanglement of (b0, b1). Similar processing is performed in the subsequent steps shown in Figure 7, and finally, all quantum entanglements are consumed and the data is placed in the original child node.

[0043] Of the above-mentioned S1 to S4, the time required for the physical operation of S2 and S3 is defined as τ, and the time required to re-establish quantum entanglement between the used communication paths is defined as T. In this case, all communication paths are used every time τn, so if n is made large enough so that τn≧T, there will be no delay in calculation due to the lack of quantum entanglement between the communication paths.

[0044] The above protocol assumes that the computing device has only two ring nodes. To perform distributed quantum computing using more ring nodes, the ring node must be connected to multiple ring nodes. This is possible by connecting each child node of a ring node to a different ring node. However, doing so would change the position of the child node where the data is placed each time the communication destination ring node changes. Therefore, when communication is desired between two ring nodes, the data must be moved to the index of a node that can communicate with each ring node. In this case, a lot of quantum entanglement is consumed within the ring node, so performance does not scale ideally as the number of communication destinations for the ring node increases. The protocol shown in the following Example 2 solves this problem.

[0045] Example 2 A description will be given of Example 2. Example 2 is an example of distributed quantum computing with nodes having four communication paths.

[0046] In the protocol of the second embodiment, each node can have four communication paths. In the second embodiment, each node is represented by a ring node consisting of n child nodes. As in the first embodiment, inside the ring node, (a i ,a i+1 ) and (b i ,b i+1 ) are connected by a communication path. As a result, two of the four communication paths that each node has are consumed by connections within the ring node.

[0047] Consider communication between two ring nodes a and b. Of the n child nodes of each ring node, m child nodes are used for communication between a and b. Let A = (α0, ...α m-1 ) and B = (β0,…β m-1 ) where A is the set of ring nodes on the a side, and B is the set of ring nodes on the b side.

[0048] In this protocol, two ring nodes are not directly connected, but rather a node string called an intermediate node layer is inserted between the two ring nodes. Each intermediate node layer consists of m node strings. There are L=ceil(log3m) intermediate node layers. In the framework described below, this intermediate node layer plays the role of routing, enabling communication regardless of whether the node where the data is located is a child node of A or B, among ring nodes a and b.

[0049] Let L number of intermediate node layers be j=0...L-1 in order from the closest to ring node a. In this case, the kth child node of the jth layer is c j,k In the second embodiment, communication paths are constructed between the ring nodes and the intermediate node layer, and between the intermediate node layers, as follows. An example of a network constructed according to the following rules in the case of n=m=4 is shown in FIG.

[0050] (1) Connection between ring nodes and intermediate nodes For all i=0...m-1, (α i ,c 0,i ) communication channel and (c L-1 ,β i ) communication path (Connections within each intermediate node layer) In each j(=0…L-1)th layer, for the k(=0…m-1)th child node, s=3 j As, (c j,k ,c j,k+s ) is connected, and (c j,k ,c j,k+m-s ) are connected. However, the subscripts within the sheaf are identified modulo m, that is, c j , k+m =c j,k It is considered to be.

[0051] In the example of FIG. 9, for example, c 0,0 and c 0,0+1 and are connected, and c 0,0 and c 0,3 (=c 0,0+4-1 ) are connected. For example, 1,0 and c1,0+3 and are connected, and c 1,0 and c 1,1 (=c 1,0+4-3 ) are connected.

[0052] (Connection between intermediate node layers) For each j=0…L-2th layer, for k=0…m-1th node, (c j,k ,c j+1,k ) communication path is created.

[0053] In each j=1...L-1th layer, for the k=0...m-1th node, (c j,k ,c j-1,k ) communication path is created.

[0054] In the example of FIG. 9, for example, c 0,0 and c 1,0 and are connected, and c 1,0 and c 0,0 and are connected.

[0055] The two ring nodes and intermediate node layer under the above conditions satisfy the following properties:

[0056] (Property) For any integer u between 0 and m, there are m pairs of start and end points {(α i ,β i+u )} i=0 m-1 When considering the set of paths connecting each of p0...p, there exists a set of paths that is an edge disjoint path set. m-1} is an edge disjoint path if any pair of paths (p i ,p j ), this means that no two paths share any edges in their paths, i.e., any edge is used at most once in all paths.

[0057] The above is any number between 0 and 3. L The integer u below can be derived from the fact that it can be written in the following form:

[0058]

Number

[0059] S1: The computing device sets j ← 0, k = i, and adds (α i , c j,k ) to the path.

[0060] S2: The computing device, if γ j is -1, adds (c j,k , c j,k+m-s ) to the path and sets k ← k + m - s.

[0061] S3: The computing device, if γ​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​An example of a case where the calculation device determines a path when n=m=4, s=2 is shown in Figure 10. As shown in Figure 10, in this example, L=2 and u=2. 2=(-1)×3 0 +(+1)×3 1 +(0)×3 2 where γ0=-1, γ1=+1, and γ2=0.

[0065] In the example of FIG. 10, the computing device creates a path connecting α3 and β1. First, the path is empty in the initial state. The computing device adds (α3, c 0,3 ) is added. Since γ0=-1, the computing device adds (c 0,3 ,c 0,3+4-1 ) and set k←k+ms. Note that c 0,3+4-1 is c 0,2 and s=3 0 Since 3+4-1=6, let k=2.

[0066] Next, the computing device adds (c 0,2 ,c 0+1,2 ) and set j←j+1, i.e., set j=1.

[0067] Next, since γ1=+1, the calculation device adds (c 1,2 ,c 1,2+3 ) and set k←k+s. In other words, here, 2+3 1 Since is 1, k=1. Since j=L-1=1, the calculation device adds (c 1,1 ,β1) is added.

[0068] From the above properties, two ring nodes connected in the middle layer can execute the following protocol. Suppose that multiple communications between ring nodes a and b are required during the calculation. At this time, the location of data within the ring node is i a ,i b Let it be the th child node.

[0069] S1: The computing device, for each ring node, determines the child node to be referenced by i a ,ib Move the node in clockwise direction until it reaches the first child node included in A and B. The child node of the destination is α i_a ∈A,β i_b Let ∈B.

[0070] S2: The computing device (α i_a ,β i_b ) communicate by consuming the quantum entanglement on the intermediate layer path between them, and set e←e+1. When the required amount of communication has been performed (when e reaches the required number of communications), the protocol is terminated.

[0071] S3: The computing device consumes the quantum entanglement on the ring node, (α i_a ,α i_a+1 ) data via α i_a From α i_a+1 Also, (β i_b ,β i_b+1 ) data via the route β i_b From β i_b+1 Move to i a ←i a +1, i b ←i b +1. Processing returns to S2.

[0072] If the time required for the S2 and S3 procedures is τ, then each route is used once in the time τm. Therefore, if the time required to generate quantum entanglement is T, then if m is set so that τm ≥ T, then the generation of quantum entanglement will not become a bottleneck. Therefore, n should be set large enough so that a sufficient number of m can be set between the ring nodes that can be connected.

[0073] <Example of operation> In the second embodiment, the calculation procedure using the above process will be described with reference to the flowchart in Fig. 11. In the following process, the calculation device generates quantum entanglement as necessary.

[0074] In S101, the calculation device obtains the positional deviation u between two ring nodes. In S102, the calculation device decomposes u into the form of the above-mentioned "Equation 1" and obtains {γ i In step S103, the process moves clockwise through the child nodes to be referenced in each ring node to a child node that has a connection between the target ring nodes.

[0075] In S104, the computing device i Communication is performed between ring nodes by executing entanglement swapping and quantum teleportation on the route calculated based on}. In S105, if the desired number of calculations have been completed, the process ends, and if not, the process returns to S103.

[0076] A specific example of the communication process in the above steps will be described with reference to FIGS.

[0077] First, assume that a communication path has been constructed between nodes in a computing device using the method described above, as shown in Fig. 12. As shown in Fig. 13, the computing device communicates between (α3, β1) by consuming the quantum entanglement of the path (white arrow line) created using the procedure described in Fig. 10.

[0078] In Fig. 14, the computing device moves data from α3 to α2 via the (α3, α2) route by consuming quantum entanglement on each ring node, and moves data from β1 to β0 via the (β1, β0) route. In Fig. 15, the computing device communicates by consuming quantum entanglement on the intermediate layer route between (α2, β0).

[0079] In Fig. 16, the computing device moves data from α2 to α1 via the (α2,α1) route by consuming quantum entanglement on each ring node, and moves data from β0 to β3 via the (β0,β3) route. In Fig. 17, the computing device communicates between (α1,β3) by consuming quantum entanglement on the intermediate layer route.

[0080] In Fig. 18, the computing device moves data from α1 to α0 via the (α1, α0) route by consuming quantum entanglement on each ring node, and moves data from β3 to β2 via the (β3, β2) route. In Fig. 19, the computing device communicates by consuming quantum entanglement on the intermediate layer route between (α0, β2).

[0081] In FIG. 20, the computing device moves data from α0 to α3 via the route (α0, α3) and from β2 to β1 via the route (β2, β1) by consuming quantum entanglement on each ring node.

[0082] (Example of hardware configuration) As described above, when a classical computer is used as the computing device (quantum network simulator), control device 2, or control device 110, the device (computing device (quantum network simulator), control device 2, or control device 110) can be realized by having the classical computer execute a program. This classical computer may be a physical computer or a virtual machine on the cloud. Hereinafter, the classical computer will be referred to as a "computer."

[0083] That is, the device can be realized by executing a program corresponding to the processing performed by the device using hardware resources such as a CPU and memory built into a computer. The program can be recorded on a computer-readable recording medium (such as a portable memory) and stored or distributed. The program can also be provided via a network such as the Internet or email.

[0084] Fig. 21 is a diagram showing an example of the hardware configuration of the computer. The computer in Fig. 21 includes a drive device 1000, an auxiliary storage device 1002, a memory device 1003, a CPU 1004, an interface device 1005, a display device 1006, an input device 1007, an output device 1008, and the like, all of which are interconnected via a bus B. The computer may further include a GPU.

[0085] A program for realizing processing on the computer is provided by a recording medium 1001 such as a CD-ROM or a memory card. When the recording medium 1001 storing the program is set in the drive device 1000, the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000. However, the program does not necessarily have to be installed from the recording medium 1001, but may be downloaded from another computer via a network. The auxiliary storage device 1002 stores the installed program as well as necessary files, data, etc.

[0086] The memory device 1003 reads and stores a program from the auxiliary storage device 1002 when an instruction to start the program is received. The CPU 1004 realizes the functions related to the device in accordance with the program stored in the memory device 1003. The interface device 1005 is used as an interface for connecting to a network, etc. The display device 1006 displays a GUI (Graphical User Interface) or the like according to the program. The input device 1007 is composed of a keyboard, mouse, buttons, a touch panel, etc., and is used to input various operation instructions. The output device 1008 outputs the results of calculations.

[0087] (Summary of implementation form, effects, etc.) According to the technology of this embodiment described above, by representing a single node with multiple "nodes arranged in a ring", the effective bandwidth between nodes is improved by increasing the number of nodes.

[0088] Distributed computing using ordinary computers also has the problem of bandwidth between nodes, but while ordinary computers can copy data, they cannot save communication resources like communication using quantum entanglement.The technology according to this embodiment utilizes the characteristic that communication bandwidth can be saved in advance due to the generation of quantum entanglement, under the quantum-specific constraint that states cannot be copied, and shows that the bandwidth in distributed quantum computing can be solved by increasing the number of nodes.

[0089] In other words, according to the technology of this embodiment, the bandwidth between two nodes can be improved by increasing the number of nodes, and the bottleneck of distributed quantum computing can be eliminated while trading off with the number of nodes.

[0090] The following additional notes are provided regarding the above-described embodiments.

[0091] <Additional Notes> (Additional note 1) a first ring node in which a plurality of nodes are connected in a ring shape; a second ring node in which a plurality of nodes are connected in a ring shape; Communicating between a node in the first ring node and a node in the second ring node by consuming quantum entanglement between the node in the first ring node and the node in the second ring node. computing device. (Additional note 2) After the communication is performed, data of the node in the first ring node is transferred to another node in the first ring node by consuming quantum entanglement between nodes in the first ring node, and data of the node in the second ring node is transferred to another node in the second ring node by consuming quantum entanglement between nodes in the second ring node. Item 1. The computing device of item 1. (Additional note 3) At least one intermediate node layer consisting of a plurality of nodes is provided between the first ring node and the second ring node; Communication is performed between the node in the first ring node and the node in the second ring node by performing entanglement swapping on a path connecting the node in the first ring node, the node in the intermediate node layer, and the node in the second ring node. Item 1. The computing device of item 1. (Additional note 4) A set of paths connecting the first ring node, the intermediate node layer, and the second ring node is an edge disjoint path set. Item 3. The computing device of item 3.

[0092] Although the present embodiment has been described above, the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims. [Explanation of symbols]

[0093] 1. Quantum Network 2. Control device 100 nodes 110 Control device 120 quantum processors 1000 Drive Device 1001 Recording media 1002 Auxiliary storage device 1003 Memory device 1004 CPU 1005 Interface device 1006 Display device 1007 Input Device 1008 Output Device

Claims

1. a first ring node in which a plurality of nodes are connected in a ring shape; a second ring node in which a plurality of nodes are connected in a ring shape; Communicating between a node in the first ring node and a node in the second ring node by consuming quantum entanglement between the node in the first ring node and the node in the second ring node. computing device.

2. After the communication is performed, data of the node in the first ring node is transferred to another node in the first ring node by consuming quantum entanglement between nodes in the first ring node, and data of the node in the second ring node is transferred to another node in the second ring node by consuming quantum entanglement between nodes in the second ring node. The computing device of claim 1 .

3. At least one intermediate node layer consisting of a plurality of nodes is provided between the first ring node and the second ring node; Communication is performed between the node in the first ring node and the node in the second ring node by performing entanglement swapping on a path connecting the node in the first ring node, the node in the intermediate node layer, and the node in the second ring node. The computing device of claim 1 .

4. A set of paths connecting the first ring node, the intermediate node layer, and the second ring node is an edge disjoint path set.

4. The computing device of claim 3.

Images