Quantum circuit semi-cutting mechanism oriented to distributed quantum computing environment

Abstract

The invention provides a quantum circuit semi-cutting mechanism for a distributed quantum computing environment, and the mechanism comprises an isomorphic sub-circuit multiplexing mechanism which is used for recognizing isomorphic sub-circuits in a quantum circuit, and cutting the quantum circuit into the isomorphic sub-circuits, so as to facilitate the multiplexing of an execution result of the isomorphic sub-circuits; according to the quantum circuit semi-cutting mechanism, remote operation is eliminated by cutting key quantum bits in a quantum circuit and gate operation, and the classic post-processing overhead during cutting of a large quantum circuit is reduced. The mechanism solves the problems that in the prior art, isomorphic sub-circuits cannot be reused, and the number of cutting operations is large; the remote communication operation in distributed quantum computing is reduced, and the execution reliability of the quantum circuit is improved.

Description

Technical Field

[0001] The present invention relates to the field of quantum computing, and in particular to a quantum circuit cutting mechanism, and more particularly to a quantum circuit half-cutting mechanism for a distributed quantum computing environment. Background Art

[0002] Quantum computing has demonstrated powerful computing acceleration capabilities for specific tasks and can accelerate the solution of difficult problems in classical computing. Using quantum computers to solve practical problems requires a large number of reliable qubits. For example, cracking RSA-2048 requires about 4,000 logical qubits. However, due to manufacturing process limitations, it is currently impossible to integrate too many qubits on a single quantum chip. Current quantum computers have at most hundreds of qubits and cannot support the operation of large quantum circuits. Distributed quantum computing (DQC) is an effective way to achieve large-scale quantum computing. In DQC, multiple small quantum chips (quantum processing units, QPUs) work together to form a large quantum computing system. The physical qubits of each QPU can be divided into two categories: the first category is data qubits, which are used to map logical qubits in quantum circuits. The second category is communication qubits, which are used to construct EPR (Einstein-Podolsky-Rosen) pairs required for long-distance communication between QPUs. Compared with local communication operations occurring within the QPU, long-distance communication operations have higher error rates and take longer. Therefore, the number of remote communication operations should be reduced to improve the execution reliability of quantum circuits in the DQC environment.

[0003] Quantum circuit cutting is another method to achieve large-scale quantum computing. A large quantum circuit can be cut into multiple sub-circuits with fewer qubits that can be executed on the QPU. Each sub-circuit is executed separately, and the execution results of these sub-circuits are merged to obtain the execution result of the original circuit. The process of merging the execution results of sub-circuits is called "classical post-processing". Quantum circuits can be cut by qubit cutting and quantum gate cutting. Figure 1 The example of using qubit cutting operation to cut a 4-qubit GHZ (Greenberger-Horne-Zeilinger) quantum circuit into two sub-circuits is shown. Qubit cutting introduces a new qubit and moves all two-qubit gate operations after the cutting position to the new qubit for execution. Figure 2 An example of using quantum gate cutting to cut a GHZ quantum circuit is shown. Quantum gate cutting cuts a two-qubit gate operation into two single-qubit gate operations, which are executed separately in two sub-circuits. The merging overhead of the sub-circuit execution results (i.e., the classical post-processing overhead) increases exponentially with the number of cuts, so the number of quantum circuit cuts should be reduced.

[0004] Quantum circuit cutting can be used to remove the remote communication operations between QPUs in a DQC environment. The existing quantum circuit cutting mechanism has the following main problems:

[0005] (1) There is currently no efficient mechanism for multiplexing isomorphic subcircuits. If a quantum circuit is divided into multiple isomorphic subcircuits, the execution results of these isomorphic subcircuits can be reused in the classical post-processing process. Only one of the multiple isomorphic subcircuits needs to be executed, and its execution result can be reused as the execution result of other isomorphic subcircuits. There is currently no efficient mechanism for multiplexing isomorphic subcircuits, and it is impossible to detect and reuse isomorphic subcircuits in a large quantum circuit. Isomorphic subcircuits need to be executed repeatedly multiple times.

[0006] (2) The existing quantum circuit cutting mechanism always cuts a quantum circuit into multiple completely independent sub-circuits. This results in a large number of cutting operations required when cutting large quantum circuits, resulting in excessive classical post-processing overhead. The inventors observed that when cutting quantum circuits, some cutting operations are unnecessary. This is because these cutting operations can only remove a small number of long-distance communication operations, but the resulting classical post-processing overhead is too high. Therefore, these unnecessary cutting operations do not need to be performed, and there is no need to cut quantum circuits into multiple completely independent sub-circuits. Summary of the invention

[0007] In order to solve the problems existing in the above-mentioned prior art, the present invention proposes a quantum circuit half-cutting mechanism for a distributed quantum computing environment, which is used to reduce remote communication operations in a DQC environment. The present invention consists of the following two parts. (1) An isomorphic sub-circuit multiplexing mechanism, which is used to cut a quantum circuit into isomorphic sub-circuits and reuse the execution results of isomorphic sub-circuits. (2) A quantum circuit half-cutting mechanism, which eliminates remote communication operations in a DQC environment by cutting key quantum bits and gate operations in the quantum circuit without cutting the quantum circuit into multiple completely independent sub-circuits, thereby reducing the classical post-processing overhead when cutting large quantum circuits. The specific steps of each part are as follows:

[0008] (I) Homogeneous sub-line multiplexing mechanism

[0009] Step 1: Get the large quantum circuit to be cut and construct an interaction graph. A node in the interaction graph represents a two-qubit gate operation that uses a logical qubit. An edge in the interaction graph represents a two-qubit gate operation, or a logical qubit shared by two adjacent two-qubit gate operations. Cutting an edge in the interaction graph is equivalent to performing a cutting operation on the quantum circuit.

[0010] Step 2: Detect isomorphic subcircuits in large quantum circuits and construct an isomorphic perception interaction graph. This includes the following steps:

[0011] Step 2.1, identify isomorphic subcircuits. Use the VF2++ algorithm to obtain isomorphic subcircuit matching solutions without quantum gate operation intersection in large quantum circuits. To avoid cross-QPU mapping of cut subcircuits, the number of qubits of each isomorphic subcircuit must not exceed the number of qubits of the largest QPU. Select the solution with the least number of cuts required to cut large quantum circuits into isomorphic subcircuits as the final solution.

[0012] Step 2.2: In the interaction graph, shrink and merge the nodes corresponding to the isomorphic subcircuits to construct an isomorphism-aware interaction graph. For each isomorphic subcircuits, shrink all the nodes corresponding to the subcircuits in the interaction graph into a supernode. Supernodes cannot be cut. Therefore, when the circuit is cut, the qubits and gate operations in the isomorphic subcircuits will not be cut, which helps to cut large quantum circuits into isomorphic subcircuits. In addition, by shrinking and merging the nodes corresponding to the isomorphic subcircuits, the number of edges in the interaction graph is reduced, the search space of the cutting scheme is reduced, and the search speed is improved.

[0013] (II) Quantum Circuit Half-cut Mechanism

[0014] Step 3: Construct a small root heap to store candidate cutting solutions. The small root heap can return the cutting solution with the lowest cutting cost among the candidate cutting solutions. The lower the cutting cost, the lower the classical post-processing overhead of the solution, and the least remote communication operations are generated after mapping to DQC. Use a Boolean list s = {s1, s2, ..., s n} represents a candidate cutting solution, where s i A value of true indicates that the i-th edge in the interaction graph is cut, s i The value false indicates that the i-th edge in the interaction graph has not been cut. The length of s represents the number of edges in the interaction graph for which a cutting decision has been made. An empty list (i.e., candidate cutting solutions that have not made a cutting decision for any edge in the interaction graph) is added to the small root heap as the initial solution.

[0015] Step 4: Iterate and update the minimum cost cutting solution until the small root pile storing the candidate cutting solutions is empty. Each iteration includes the following steps:

[0016] Step 4.1: Take out a candidate cutting solution from the small root pile storing the candidate cutting solutions.

[0017] Step 4.2: Check whether the candidate cutting scheme is feasible. If the candidate cutting scheme is not feasible, skip the scheme and proceed to the next iteration. The following three types of candidate cutting schemes are not feasible: (1) If the candidate cutting scheme is used to cut the circuit, resulting in the number of quantum bits of a sub-circuit exceeding the number of quantum bits of the maximum QPU, the candidate cutting scheme is not feasible. (2) If the cutting cost of the candidate cutting scheme exceeds the minimum cost cutting scheme, the candidate cutting scheme is not feasible. This is because further iterations will not produce a better solution than the minimum cost cutting scheme. (3) If the candidate cutting scheme has made cutting decisions for all edges in the interaction graph (that is, the length of the Boolean value list of the candidate cutting scheme is equal to the number of edges in the interaction graph), but the large quantum circuit has not been cut into completely independent sub-circuits, the candidate cutting scheme is not feasible.

[0018] Step 4.3, update the minimum cost cutting plan. When the candidate cutting plan is feasible and the candidate cutting plan has made cutting decisions for all edges in the interaction graph, compare the candidate cutting plan with the minimum cost cutting plan. If the cutting cost of the candidate cutting plan is lower, set the candidate cutting plan as the new minimum cost cutting plan. The cutting cost is represented by a tuple containing four values. The first item in the tuple is the number of remote communication operations generated when mapping the cut sub-circuit to DQC. The second item is the sampling overhead (that is, the number of times the sub-circuit needs to be executed), which is calculated as O(16 k1 9 k2 ), where k1 represents the number of qubit cuts and k2 represents the number of quantum gate cuts. If there are reusable isomorphic subcircuits, the sampling overhead will be reduced. The third term is the classical post-processing overhead, calculated as O(4 k1 6 k2 ). The last term is -1×search depth (i.e., the length of candidate cut s). In the min-heap, candidate cuts are sorted by the lower-indexed items in the tuple. If the lower-indexed items in two tuples are equal, the higher-indexed items are compared.

[0019] Step 4.4, construct a new candidate cutting scheme. For a feasible candidate cutting scheme, if it does not make cutting decisions for all edges in the interaction graph, add a true to the candidate cutting scheme to generate the first new candidate cutting scheme; add a false to the candidate cutting scheme to generate the second new candidate cutting scheme. Add the two new candidate cutting schemes to the small root heap storing candidate cutting schemes.

[0020] Step 4.5: If the small root heap storing candidate cutting solutions is not empty, repeat steps 4.1 to 4.4. Otherwise, return the lowest cost cutting solution.

[0021] Step 5: Remove non-critical cutting operations in the minimum cost cutting scheme and construct a half-cutting scheme. For each cutting operation in the minimum cost cutting scheme, obtain the number of remote communication operations that can be removed by executing the cutting operation. Calculate the average number of remote communication operations that can be removed by all cutting operations. If the number of remote communication operations that can be removed by a cutting operation is greater than or equal to the average, then the cutting operation is recorded as a critical cutting operation, otherwise it is recorded as a non-critical cutting operation. Output a half-cutting scheme that only contains critical cutting operations.

[0022] The present invention also proposes a storage medium for storing a program for executing any one of the quantum circuit half-cut mechanisms for a distributed quantum computing environment.

[0023] The advantages of the present invention are as follows: the present invention designs an isomorphic subcircuit multiplexing mechanism and a quantum circuit half-cutting mechanism for the problem of quantum circuit cutting in a distributed quantum computing environment. The isomorphic subcircuit multiplexing mechanism can identify isomorphic subcircuits in quantum circuits, construct isomorphic perception interaction graphs, and cut quantum circuits into isomorphic subcircuits, thereby reducing the number of subcircuits that need to be executed. In the quantum circuit half-cutting mechanism, only key quantum bits and gate operations are cut, reducing the number of cutting operations required to cut large quantum circuits and reducing classical post-processing overhead. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of the quantum bit cutting operation performed on the GHZ quantum circuit;

[0025] Figure 2 Schematic diagram of quantum gate cutting operation performed on GHZ quantum circuit;

[0026] Figure 3 A flow chart of a quantum circuit half-cut mechanism for a distributed quantum computing environment proposed by the present invention;

[0027] Figure 4 An interaction diagram constructed for a large quantum circuit in an embodiment of the present invention;

[0028] Figure 5 To identify isomorphic subcircuits, a schematic diagram of the isomorphic perceptual interaction graph is constructed;

[0029] Figure 6 A schematic diagram for updating the minimum cost cutting scheme;

[0030] Figure 7 To remove non-critical cutting operations in the minimum cost cutting scheme, a schematic diagram of the half-cutting scheme is constructed. Specific implementation methods

[0032] In order to make the above features and effects of the present invention more clearly understood, the following embodiments are given in conjunction with the accompanying drawings ( Figures 1 to 7 ) are described in detail as follows.

[0033] A quantum circuit half-cutting mechanism for a distributed quantum computing environment comprises the following steps: step 1, based on an isomorphic subcircuit reuse mechanism, obtaining a large quantum circuit to be cut and constructing an interaction graph; step 2, based on the isomorphic subcircuit reuse mechanism, detecting isomorphic subcircuits in the large quantum circuit and constructing an isomorphism-aware interaction graph; step 3, constructing a small root pile for storing candidate cutting schemes; step 4, iteratively updating the minimum cost cutting scheme until the small root pile for storing the candidate cutting schemes is empty; step 5, removing non-critical cutting operations in the minimum cost cutting scheme and constructing a half-cutting scheme, without cutting the quantum circuit into multiple completely independent subcircuits, eliminating remote communication operations in a DQC environment by only cutting key quantum bits and gate operations in the quantum circuit, and reducing the classical post-processing overhead when cutting large quantum circuits.

[0034] Step 1 includes: a node in the interaction graph represents a two-qubit gate operation using a logical qubit; an edge in the interaction graph represents a two-qubit gate operation, or a logical qubit shared by two adjacent two-qubit gate operations; cutting an edge in the interaction graph is equivalent to performing a cutting operation on the quantum circuit.

[0035] Step 2 includes: Step 2.1, identifying isomorphic subcircuits: using the VF2++ algorithm to obtain isomorphic subcircuit matching solutions without quantum gate operation intersections in large quantum circuits; to avoid cross-QPU mapping of the cut subcircuits, the number of qubits of each isomorphic subcircuit must not exceed the number of qubits of the largest QPU; selecting the solution with the least number of cuts required to cut the large quantum circuit into isomorphic subcircuits as the final solution; Step 2.2, shrinking the nodes corresponding to the isomorphic subcircuits in the interaction graph to construct an isomorphic perception interaction graph. For each isomorphic subcircuit, all nodes corresponding to the subcircuit in the interaction graph are shrunk into a supernode.

[0036] The present invention provides a quantum circuit half-cutting mechanism for a distributed quantum computing environment, including: an isomorphic subcircuit multiplexing mechanism for identifying isomorphic subcircuits in a quantum circuit and cutting the quantum circuit into isomorphic subcircuits to facilitate multiplexing of the execution results of the isomorphic subcircuits; a quantum circuit half-cutting mechanism for eliminating remote operations by cutting key quantum bits and gate operations in the quantum circuit, thereby reducing the classical post-processing overhead when cutting large quantum circuits. This mechanism solves the problem of the inability to reuse isomorphic subcircuits and the large number of cutting operations in the prior art; it reduces remote communication operations in distributed quantum computing and improves the execution reliability of quantum circuits.

[0037] The present invention consists of the following two parts: (1) a homogeneous subcircuit multiplexing mechanism, which is used to cut quantum circuits into homogeneous subcircuits and reuse the execution results of homogeneous subcircuits. (2) a quantum circuit half-cutting mechanism, which eliminates the remote communication operations in the DQC environment by cutting the key qubits and gate operations in the quantum circuit without cutting the quantum circuit into multiple completely independent subcircuits, thereby reducing the classical post-processing overhead when cutting large quantum circuits.

[0038] 1. Operating environment. The present invention is based on the open source software development toolkit Qiskit and Circuit knitting toolbox to implement the quantum circuit splitting mechanism in the distributed quantum computing system. The interface provided by Qiskit is used to obtain the interaction information between large quantum circuits and their qubits, and an interaction graph is constructed. The locally constructed quantum circuit cutting mechanism is used to cut large quantum circuits, and the split sub-circuits are run using Circuit_knitting_toolbox and Qiskit, and the execution results of the sub-circuits are merged to obtain the execution results of the original circuits.

[0039] 2. Specific steps. Figure 3 A flow chart of an embodiment of the present invention is shown. The present invention mainly includes the following steps:

[0040] (I) Homogeneous sub-line multiplexing mechanism

[0041] Step 1: Obtain the large quantum circuit to be cut and construct an interaction diagram. Figure 4 An example of constructing an interaction graph for a large quantum circuit is shown. The method of constructing the interaction graph is as follows: for each two-qubit gate operation, two nodes are added to the interaction graph, representing the two logical qubits used by the two-qubit gate operation. For two nodes in the interaction graph that belong to the same two-qubit gate operation, there is an edge between the nodes, representing the two-qubit gate operation, that is, Figure 4 Cutting a horizontal solid line edge in the interaction graph is equivalent to performing a quantum bit cut on the quantum circuit. Figure 1 An example of using a qubit cut operation to cut a 4-qubit GHZ quantum circuit is shown. If two adjacent two-qubit gate operations use the same logical qubit, there is an edge between the corresponding nodes, representing the logical qubit shared by the two two-qubit gate operations, that is, Figure 4 Cutting a vertical dashed edge in the interaction graph is equivalent to performing a quantum gate cut on the quantum circuit. Figure 2 An example of cutting a 4-qubit GHZ quantum circuit using quantum gate cutting operations is demonstrated.

[0042] Step 2: Detect isomorphic subcircuits in large quantum circuits and construct an isomorphic perception interaction graph. This step includes:

[0043] Step 2.1: Identify isomorphic subcircuits. Use the VF2++ algorithm to obtain isomorphic subcircuit matching solutions with no intersection of quantum gate operations in large quantum circuits. Figure 5 - (a) shows an example of identifying isomorphic subcircuits. The identified isomorphic subcircuits are marked with grey dashed boxes.

[0044] Step 2.2: In the interaction graph, shrink and merge the nodes corresponding to the isomorphic sub-circuits to construct an isomorphism-aware interaction graph. Figure 5 -(b) shows an example of constructing an isomorphism-aware interaction graph. Figure 5 -For each isomorphic subcircuit in (a), all nodes corresponding to the subcircuit in the interaction graph are collapsed into a supernode to construct an isomorphism-aware interaction graph. Supernodes cannot be further cut, that is, the qubits and gate operations in the isomorphic subcircuit will not be cut, which helps to cut large quantum circuits into isomorphic subcircuits.

[0045] (II) Quantum Circuit Half-cut Mechanism

[0046] Step 3: Build a small root heap to store candidate cutting solutions.

[0047] Step 4: Iterate and update the minimum cost cutting solution until the small root pile storing the candidate cutting solutions is empty. Each iteration includes the following steps:

[0048] Step 4.1: Take out a candidate cutting solution from the small root pile storing the candidate cutting solutions.

[0049] Step 4.2: Check whether the candidate cutting solution is feasible. If the candidate cutting solution is not feasible, skip the solution and proceed to the next iteration.

[0050] Step 4.3: Update the minimum cost cutting plan. Figure 6 An example of updating the minimum cost cutting plan is shown. Figure 6 - (a) shows a 4-qubit quantum supremacy circuit and its interaction diagram. If each QPU in DQC has only 3 qubits, the circuit needs to be mapped across QPUs to generate long-distance communication operations. Figure 6-(b) shows the current minimum cost cut. This scheme uses two qubit cut operations to cut the line into 4-qubit sub-circuit 1 and 2-qubit sub-circuit 2. Mapping sub-circuit 1 requires at least one remote communication operation, so the first term of the cut cost of the minimum cost cut is 1. The candidate line cut obtained from the minimum heap cuts the line into two 3-qubit sub-circuits, each of which can be mapped to a QPU without any remote communication operation. Therefore, the first term of the cut cost is 0. In contrast, the candidate line cut obtained from the minimum heap has a lower cutting cost, so this scheme is set as the new minimum cost cut.

[0051] Step 4.4: Construct a new candidate cutting scheme.

[0052] Step 4.5: Return the lowest cost cutting solution.

[0053] Step 5: Remove non-critical cutting operations in the minimum cost cutting solution and construct a half-cutting solution. Figure 7 The present invention shows an example of removing non-critical operations from the minimum cost cutting scheme and constructing a half-cut scheme. If a 4-qubit quantum Fourier transform (QFT) circuit is to be mapped to a DQC environment with only a 3-qubit QPU, at least six remote operations need to be performed without cutting the quantum circuit. The minimum cost cutting scheme returned in step 4.5 is as follows: Figure 7 As shown in (a), the scheme uses two quantum bit cutting operations and two quantum gate cutting operations. Figure 7 -(b) shows the line generated after using only cutting operation ①. At least four remote operations are required to map this line. Therefore, the number of remote gate operations that can be removed using cutting operation ① is 6-4=2. Similarly, using only cutting operations ②, ③, and ④, the number of remote operations that can be removed are 2, 1, and 1 respectively. The average value of the remote operations that can be removed by all cutting operations is (2+2+1+1) / 4=1.5. When the number of remote operations that can be removed by a cutting operation is greater than or equal to the average value, the cutting operation is recorded as a critical cutting operation. Figure 7 - In (c), cutting operations ① and ② are critical cutting operations. Otherwise, the cutting operation is recorded as a non-critical cutting operation. Figure 7 - In (c), the cutting operations ③ and ④ are recorded as non-critical cutting operations. The output only contains the half-cut scheme of the critical cutting operations ① and ②. This scheme does not cut the quantum circuit into multiple completely independent sub-circuits.

[0054] The present invention also proposes a storage medium for storing a program for executing any one of the quantum circuit half-cut mechanisms for a distributed quantum computing environment.

Claims

1. A quantum circuit half-cut mechanism for a distributed quantum computing environment, characterized in that: The following steps are involved: Step 1: Based on the isomorphic sub-circuit multiplexing mechanism, obtain the large quantum circuit to be cut and construct an interaction graph; Step 2: Based on the isomorphic subcircuit reuse mechanism, detect isomorphic subcircuits in large quantum circuits and construct an isomorphic perception interaction graph; Step 3, construct a small root heap to store candidate cutting plans; Step 4, iteratively update the minimum cost cutting plan until the small root heap storing the candidate cutting plans is empty; Step 5, remove non-critical cutting operations in the minimum cost cutting scheme and construct a half-cutting scheme. There is no need to cut the quantum circuit into multiple completely independent sub-circuits. Only by cutting the key qubits and gate operations in the quantum circuit, the remote communication operations in the DQC environment are eliminated, reducing the classical post-processing overhead when cutting large quantum circuits.

2. A quantum circuit half-cut mechanism for a distributed quantum computing environment as claimed in claim 1, characterized in that: Step 1 includes: a node in the interaction graph represents a two-qubit gate operation using a logical qubit; an edge in the interaction graph represents a two-qubit gate operation, or a logical qubit shared by two adjacent two-qubit gate operations; cutting an edge in the interaction graph is equivalent to performing a cutting operation on the quantum circuit.

3. A quantum circuit half-cut mechanism for a distributed quantum computing environment as claimed in claim 1, characterized in that: Step 2 includes: Step 2.1, identify isomorphic subcircuits: use the VF2++ algorithm to obtain isomorphic subcircuit matching solutions without quantum gate operation intersection in large quantum circuits; to avoid cross-QPU mapping of cut subcircuits, the number of qubits of each isomorphic subcircuit must not exceed the number of qubits of the largest QPU; select the solution with the least number of cuts required to cut large quantum circuits into isomorphic subcircuits as the final solution; Step 2.2, in the interaction graph, the nodes corresponding to the isomorphic sub-circuits are collapsed and merged to construct an isomorphism-aware interaction graph. For each isomorphic sub-circuits, all nodes corresponding to the sub-circuits in the interaction graph are collapsed and merged into a supernode.

4. A quantum circuit half-cut mechanism for a distributed quantum computing environment as claimed in claim 1, characterized in that: Step 3 includes: the small root heap can return the cutting scheme with the lowest cutting cost among the candidate cutting schemes; the lower the cutting cost, the lower the classical post-processing overhead of the scheme, and the least remote communication operations are generated after mapping to DQC; the Boolean value list s = {s1, s2, ..., s n } represents a candidate cutting solution, where s i A value of true indicates that the i-th edge in the interaction graph is cut, s i A value of false indicates that the i-th edge in the interaction graph has not been cut; the length of s represents the number of edges in the interaction graph for which cutting decisions have been made; an empty list, that is, candidate cutting plans for which no cutting decisions have been made for any edge in the interaction graph, is added to the small root heap as the initial plan.

5. A quantum circuit half-cut mechanism for a distributed quantum computing environment as claimed in claim 1, characterized in that: Step 4 includes: Step 4.1, taking out a candidate cutting solution from the small root pile storing the candidate cutting solutions; Step 4.2, check whether the candidate cutting scheme is feasible: if the candidate cutting scheme is not feasible, skip the scheme and continue to the next iteration; the following three types of candidate cutting schemes are not feasible: (1) If the candidate cutting scheme is used to cut the line, resulting in the number of quantum bits of a sub-circuit exceeding the number of quantum bits of the maximum QPU, the candidate cutting scheme is not feasible; (2) If the cutting cost of the candidate cutting scheme exceeds the minimum cost cutting scheme, the candidate cutting scheme is not feasible because further iterations will not produce a better solution than the minimum cost cutting scheme; (3) If the candidate cutting scheme has made cutting decisions for all edges in the interaction graph, that is, the length of the Boolean value list of the candidate cutting scheme is equal to the number of edges in the interaction graph, but the large quantum circuit has not been cut into completely independent sub-circuits, then the candidate cutting scheme is not feasible; Step 4.3, update the minimum cost cutting plan: when the candidate cutting plan is feasible and the candidate cutting plan has made cutting decisions for all edges in the interaction graph, compare the candidate cutting plan with the minimum cost cutting plan; if the cutting cost of the candidate cutting plan is lower, set the candidate cutting plan as the new minimum cost cutting plan; the cutting cost is represented by a tuple containing four values; the first item in the tuple is the number of remote communication operations generated when mapping the cut sub-circuit to DQC; the second item is the sampling overhead, that is, the number of times the sub-circuit needs to be executed, which is calculated as O(16 k1 9 k2 ), where k1 represents the number of quantum bit cuts and k2 represents the number of quantum gate cuts; if there are reusable isomorphic subcircuits, the sampling overhead will be reduced; the third term is the classical post-processing overhead, calculated as O(4 k1 6 k2 ); the last item is -1×search depth, which is the length of candidate cut s; in the minimum heap, the candidate cuts are sorted according to the items with lower index values ​​in the tuple; if the items with lower index values ​​in two tuples are equal, the items with higher index values ​​are compared; Step 4.4, construct a new candidate cutting plan: for a feasible candidate cutting plan, if it does not make cutting decisions for all the edges in the interaction graph, add a true to the candidate cutting plan to generate the first new candidate cutting plan; add a false to the candidate cutting plan to generate the second new candidate cutting plan; add the two new candidate cutting plans to the small root heap storing the candidate cutting plans; Step 4.5, if the small root heap storing the candidate cutting solutions is not empty, repeat steps 4.1 to 4.4, otherwise return the lowest cost cutting solution.

6. A quantum circuit half-cutting mechanism for a distributed quantum computing environment as claimed in claim 1, characterized in that: Step 5 includes: for each cutting operation in the minimum cost cutting scheme, obtaining the number of remote communication operations that can be removed by executing the cutting operation; calculating the average number of remote communication operations that can be removed by all cutting operations; if the number of remote communication operations that can be removed by a cutting operation is greater than or equal to the average value, then the cutting operation is recorded as a critical cutting operation, otherwise it is recorded as a non-critical cutting operation. Output a half-cutting scheme that only contains critical cutting operations.

7. A storage medium for storing a program for executing a quantum circuit half-cut mechanism for a distributed quantum computing environment as claimed in any one of claims 1 to 6.

Images