CN115907022A - A multi-quantum service conversion and simulation scheduling method, device, equipment and medium - Google Patents

Abstract

The invention provides a method, a device, equipment and a readable medium for converting and simulating multi-quantum services, wherein the method comprises the following steps: collecting data of each quantum system unit through a cloud data center, and constructing a quantum business system simulation environment based on the collected data; designing a quantum business system dispatcher decision model; obtaining data of a quantum business system dispatcher executing a decision, and training a quantum business system dispatcher decision model in a simulation environment based on the obtained data of the executing decision; and using the trained quantum business system dispatcher decision model to assist a quantum business system dispatcher in executing a decision. By using the scheme of the invention, the influence caused by artificial misoperation can be reduced, the scheduling efficiency and accuracy are improved, the difficulty of micro-service application deployment and management can be reduced, and the adaptive management of different types of quantum services can be realized. The present invention relates to the field of computers.

Description

A multi-quantum service conversion and simulation scheduling method, device, equipment and medium

technical field

The present invention relates to the field of computers, and more specifically relates to a method, device, equipment and readable medium for multi-quantum service conversion and simulation scheduling.

Background technique

In recent years, reinforcement learning technology has received widespread attention, especially in combination with deep learning, which has brought great progress to the field of artificial intelligence. Reinforcement learning is different from traditional supervised learning, mainly in the reinforcement signal. The reinforcement signal provided by the environment in reinforcement learning is an evaluation of the quality of the generated action (usually a scalar signal), rather than telling the reinforcement learning system RLS (reinforcement learning system) how to produce the correct action. Reinforcement learning continuously learns to make optimal actions in different environments through the task of interaction between the agent and the environment, and uses these perception generation strategies to create higher machine intelligence. Reinforcement learning has been applied in the fields of robot control, autonomous driving, and recommender systems, and has surpassed human performance in many areas.

Large-scale quantum business system cluster scheduling is a combination of computing power provided by many quantum computing participating nodes and classical computing. It is a complex large-scale cluster system. With the development of quantum computing, there will be higher requirements for quantum system clusters. Data from various computing nodes are collected through various sensors, and data analysis is used to better understand the status of participants participating in large-scale quantum systems, which can be discovered in advance How to effectively use reinforcement learning technology to simulate the real environment through the simulation of large-scale quantum business system clusters to form a large-scale quantum business system cluster dispatcher An accurate and efficient scheduling strategy to avoid adverse consequences caused by scheduling errors has become a problem to be solved.

Contents of the invention

In view of this, the purpose of the embodiments of the present invention is to propose a method, device, device and readable medium for multi-quantum service conversion and simulation scheduling. By using the technical solution of the present invention, the impact of human error can be reduced. Improving scheduling efficiency and accuracy can reduce the difficulty of micro-service application deployment and management, and can realize the adaptive management of different types of quantum services.

Based on the above purpose, an aspect of the embodiments of the present invention provides a method for multi-quantum service conversion and simulation scheduling, including the following steps:

Collect the data of each quantum system unit through the cloud data center, and build a quantum business system simulation environment based on the collected data;

Design the decision-making model of the dispatcher of the quantum business system;

Obtain the decision-making data of the dispatcher of the quantum business system, and train the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the obtained data of the decision-making execution;

Use the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher in making decisions.

According to an embodiment of the present invention, collecting data of each quantum system unit through the cloud data center, and constructing a quantum business system simulation environment based on the collected data includes:

Collect the operation data of each quantum system unit, wherein the operation data includes the real-time status data, operation log, scheduling plan, fault data and dispatcher scheduling execution data of each quantum system unit;

Based on the collected data and using digital twin technology to build a quantum business system simulation environment.

According to an embodiment of the present invention, obtaining the data of the decision-making execution of the dispatcher of the quantum business system, and training the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the data of the obtained decision-making execution include:

Obtain the data of the decision-making of the dispatcher of the quantum business system;

According to the actual operation of the quantum business system dispatcher, combined with the state environment of the quantum business system, the quantum business system dispatcher dispatches the operation instruction execution sequence;

According to the current quantum business system plan, the operating status of each quantum system unit and the practice of abnormal accidents, determine the next step to execute the scheduling operation in the simulation environment;

According to the scheduling operation of the dispatcher of the quantum business system and the actual execution effect, combined with the scheduling operation determined in the simulation environment, the reward function of the simulation environment of the quantum business system is set;

Based on the reward function and using the A3C algorithm to train the decision-making model of the quantum business system dispatcher.

According to an embodiment of the present invention, based on the reward function and adopting the A3C algorithm to train the decision-making model of the dispatcher of the quantum business system includes:

Set the number of worker threads, the number of global shared iterations, the global maximum number of iterations, the state feature dimension and the global parameters of the operation instruction set, set the global model public neural network, and set the initialization state of the simulation environment;

Initialize the decision-making model of the quantum business system dispatcher;

Make each worker thread use the global model public neural network to independently interact with the simulation environment, perform scheduling operations to obtain feedback and update the gradient of the local global model public neural network, and update the model parameters of the global model public neural network;

Repeat the previous step until the quantum business system dispatcher decision model converges.

According to an embodiment of the present invention, the global model public neural network includes an Actor network and a Critic network.

According to an embodiment of the present invention, using the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher to perform decision-making includes:

Collect the actual operating data of the current quantum business system in real time and update it to the simulation environment in real time;

Use the trained quantum business system dispatcher decision-making model to output the next step of dispatching operation according to the current actual situation;

Record the current state of the simulation environment and the next scheduling operation output by the dispatcher's decision model of the quantum business system, and feedback and update the simulation environment;

Set a time period, repeat the above steps to form the quantum business system scheduler's recommended operation sequence, and use the recommended operation sequence to assist the quantum business system scheduler in making decisions.

According to an embodiment of the present invention, using the recommended operation sequence to assist the quantum business system dispatcher to perform decision-making includes:

The dispatcher of the quantum business system refers to the recommended operation sequence and performs scheduling operations in combination with the actual situation;

After the operation is scheduled, the actual operation results are obtained, and the actual operation data of the current quantum business system is updated, and updated in the simulation environment in real time.

According to an embodiment of the present invention, also include:

Optimize the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher.

According to an embodiment of the present invention, optimizing the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher includes:

Obtain the actual operation data of each quantum business system dispatcher;

Use the actual operation data to train the decision-making model of the quantum business system dispatcher in the simulation environment;

Use the quantum business system dispatcher decision model to simulate the operation and scheduling of the quantum business system in a simulation environment;

Compare the scheduling results of the quantum business system dispatcher decision-making model with the optimal scheduling instructions;

According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized.

According to an embodiment of the present invention, also include:

Compare the dispatching operation of the dispatcher decision model of the quantum business system with the optimal dispatching instruction;

According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized.

According to an embodiment of the present invention, collecting data of each quantum system unit through the cloud data center includes:

The data of superconducting quantum system, nuclear spin quantum system, optical cavity quantum system and ion well quantum system are collected through the cloud data center.

According to an embodiment of the present invention, the cloud data center is converted from the actual quantum business system of each quantum system unit into a business that can be processed by the cloud center through the quantum business adapter.

According to an embodiment of the present invention, also include:

Create microservices in the cloud data center to encapsulate different quantum business processing classes in the cloud data center adaptation interface.

According to an embodiment of the present invention, creating a microservice in a cloud data center includes:

Obtain a microservice creation request, where the microservice creation request includes creation information and target quantum business;

Find the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business;

Use the target processing class to deploy the creation information to create an instance of the microservice on the target quantum business.

According to an embodiment of the present invention, also include:

In response to receiving the microservice creation request input by the user in the cloud data center, obtain the creation information of the creation request and the target quantum business;

Based on the target quantum business, call the corresponding processing class in the cloud data center to establish a microservice instance.

According to an embodiment of the present invention, creating a microservice in a cloud data center includes:

Obtain metadata and service type instances corresponding to various types of quantum computers;

orchestrating service type instances to build a software environment for service type instances;

Configure the parameters of the corresponding service type instance according to the metadata to obtain the processing class of each quantum business.

According to an embodiment of the present invention, searching for the target processing class corresponding to the target quantum service according to the corresponding relationship between the processing class and the quantum service includes:

Judging whether the target quantum business is recorded in the corresponding relationship between the processing class and the quantum business;

In response to recording the target quantum service, use the processing class corresponding to the target quantum service as the target processing class;

In response to no target quantum service being recorded, a pre-set general processing class is invoked as the target processing class.

Another aspect of the embodiments of the present invention also provides a device for multi-quantum service conversion and simulation scheduling, including:

Building modules, the building modules are configured to collect data of each quantum system unit through the cloud data center, and build a quantum business system simulation environment based on the collected data;

Design module, the design module is configured to design the dispatcher decision model of the quantum business system;

A training module, the training module is configured to obtain the data of the decision-making execution of the dispatcher of the quantum business system, and train the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the obtained data of the execution decision-making;

The execution module is configured to use the trained decision-making model of the quantum business system dispatcher to assist the quantum business system dispatcher to execute the decision.

Another aspect of the embodiments of the present invention also provides a computer device, the computer device includes:

at least one processor; and

Memory, the memory stores computer instructions that can be run on the processor, and when the instructions are executed by the processor, the steps of any one of the above-mentioned methods are realized.

Another aspect of the embodiments of the present invention further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps of any one of the above methods are implemented.

The present invention has the following beneficial technical effects: the method of quantum business system simulation scheduling provided by the embodiment of the present invention collects data of each quantum system unit through the cloud data center, and builds a quantum business system simulation environment based on the collected data; design The decision-making model of the quantum business system dispatcher; obtain the data of the execution decision of the quantum business system dispatcher, and train the decision-making model of the quantum business system dispatcher in the simulation environment based on the obtained execution decision-making data; use the trained quantum business system scheduling The decision-making model of the quantum business system assists the dispatcher of the quantum business system to implement the technical solution of the decision-making, which can reduce the impact of human error, improve the efficiency and accuracy of scheduling, reduce the difficulty of deploying and managing micro-service applications, and realize the control of different types of quantum Business adaptation management.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention, and those skilled in the art can obtain other embodiments according to these drawings without any creative effort.

FIG. 1 is a schematic flowchart of a method for multi-quantum service conversion and simulation scheduling according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a device for multi-quantum service conversion and simulation scheduling according to an embodiment of the present invention;

Figure 3 is a schematic diagram of a computer device according to one embodiment of the present invention;

FIG. 4 is a schematic diagram of a computer readable storage medium according to one embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

Based on the above purpose, the first aspect of the embodiments of the present invention proposes an embodiment of a method for multi-quantum service conversion and simulation scheduling. Figure 1 shows a schematic flowchart of the method.

As shown in Figure 1, the method may include the following steps:

S1 collects the data of each quantum system unit through the cloud data center, and builds a quantum business system simulation environment based on the collected data. The data from each participating quantum system unit is collected through the large-scale quantum business cluster cloud data center, and the simulation environment of the quantum business system is formed by using digital twin technology. The collected data is mainly the operation data of each quantum system unit, and the operation data includes each quantum system. Unit real-time status data, operation logs, scheduling plans, fault data, and dispatcher scheduling execution data. Each quantum system unit includes a superconducting quantum system, a nuclear spin quantum system, an optical cavity quantum system and an ion well quantum system. The simulation environment of the quantum business system runs in the cloud data center of the quantum business system. Through the data collection and instruction control of each participating quantum business system unit, the data twin operation body participating in the quantum business system unit is realized. The quantum business system cloud data center is used for Provide computing, storage and network cloud technology facilities services, and run the business management system of the quantum system, collect the sensing data of each participating quantum business system unit, and provide the computing power and storage required for intensive learning training, from each quantum participating The data of the node is the unit data of the quantum business system, and this data has been transformed into the classical data of the quantum business.

S2 Design the decision-making model of the dispatcher of the quantum business system. The reinforcement learning quantum business system dispatcher decision-making model is a strategy model formed by simulating the actual operation of the dispatcher. It is obtained through intensive training and is a serialized neural network model. It is determined according to the current business plan, business process and actual business operation of the quantum business system. The scheduling action to be performed by the system scheduler.

S3 obtains the decision-making data of the dispatcher of the quantum business system, and trains the decision-making model of the dispatcher of the quantum business system in a simulated simulation environment based on the data of the decision-making execution obtained. Firstly, obtain the decision-making data of the dispatcher of the quantum business system, construct the dispatcher’s scheduling operation command execution sequence according to the actual operation of the dispatcher of the quantum business system, and combine with the state environment of the quantum business system, according to the current quantum business system plan, each The operating status of the quantum system unit and the actual situation of abnormal accidents are determined in the simulation environment to determine the next step to execute the scheduling operation, and the quantum business is set according to the scheduling operation of the quantum business system dispatcher and the actual execution effect, combined with the scheduling operation determined in the simulation environment The reward function of the system simulation environment is based on the reward function and the A3C algorithm is used to train the decision-making model of the quantum business system dispatcher. The training process includes setting the number of worker threads, the number of global shared iterations, the global maximum number of iterations, the state feature dimension and the global parameters of the operation instruction set, and setting the global model public neural network, where the global model public neural network includes Actor network and Critic network , set the initialization state of the simulation environment, initialize the decision-making model of the quantum business system dispatcher, make each worker thread interact with the simulation environment independently by using the global model public neural network, and update the local global model public neural network after executing the scheduling operation and obtaining feedback Gradient, update the model parameters of the public neural network of the global model, and execute the previous step in a loop until the decision model of the quantum business system dispatcher converges.

S4 uses the trained quantum business system dispatcher's decision-making model to assist the quantum business system dispatcher in making decisions.

The technical solution of the present invention can reduce the impact of human error, improve scheduling efficiency and accuracy, reduce the difficulty of deploying and managing microservice applications, and realize adaptive management of different types of quantum services.

In a preferred embodiment of the present invention, collecting the data of each quantum system unit through the cloud data center, and constructing a quantum business system simulation environment based on the collected data includes:

Collect the operation data of each quantum system unit, wherein the operation data includes the real-time status data, operation log, scheduling plan, fault data and dispatcher scheduling execution data of each quantum system unit;

Based on the collected data and using digital twin technology to build a quantum business system simulation environment.

In a preferred embodiment of the present invention, obtaining the data of the execution decision-making of the quantum business system dispatcher, and training the decision-making model of the quantum business system dispatcher in a simulation environment based on the obtained data of execution decision-making includes:

Obtain the data of the decision-making of the dispatcher of the quantum business system;

According to the actual operation of the quantum business system dispatcher, combined with the state environment of the quantum business system, the quantum business system dispatcher dispatches the operation instruction execution sequence;

According to the current quantum business system plan, the operating status of each quantum system unit and the practice of abnormal accidents, determine the next step to execute the scheduling operation in the simulation environment;

According to the scheduling operation of the dispatcher of the quantum business system and the actual execution effect, combined with the scheduling operation determined in the simulation environment, the reward function of the simulation environment of the quantum business system is set;

Based on the reward function and using the A3C algorithm to train the decision-making model of the quantum business system dispatcher.

In a preferred embodiment of the present invention, based on the reward function and adopting the A3C algorithm to train the quantum business system dispatcher decision-making model includes:

Set the number of worker threads, the number of global shared iterations, the global maximum number of iterations, the state feature dimension and the global parameters of the operation instruction set, set the global model public neural network, and set the initialization state of the simulation environment;

Initialize the decision-making model of the quantum business system dispatcher;

Make each worker thread use the global model public neural network to independently interact with the simulation environment, perform scheduling operations to obtain feedback and update the gradient of the local global model public neural network, and update the model parameters of the global model public neural network;

Repeat the previous step until the quantum business system dispatcher decision model converges.

In a preferred embodiment of the present invention, the global model public neural network includes an Actor network and a Critic network. Apply for resources in the cloud data center of the quantum business system, and use the A3C algorithm to train the decision-making model of the dispatcher of the quantum business system for reinforcement learning. First, set the number of worker threads, the number of global shared iterations, the global maximum number of iterations, the dimension of state features, and the global scope of the operation instruction set Parameters, initialize the dispatcher decision-making model of the reinforcement learning quantum business system, set the global model public neural network, and set the initialization state of the quantum business system simulation environment, where the global model public neural network includes Actor network and Critic network, using the A3C algorithm to make Each worker thread adopts the Actor network and Critic network structure, independently interacts with the quantum business system simulation environment, performs scheduling operations to obtain feedback, and updates the local Actor network and Critic network gradients, and brings together the updated results to the global model public neural network network, update the model parameters of the global model public neural network, execute the above steps in a loop until the reinforcement learning quantum business system dispatcher decision model converges, and obtain the optimal reinforcement learning quantum business system dispatcher decision model.

In a preferred embodiment of the present invention, using the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher to perform decision-making includes:

Collect the actual operating data of the current quantum business system in real time and update it to the simulation environment in real time;

Use the trained quantum business system dispatcher decision-making model to output the next step of dispatching operation according to the current actual situation;

Record the current state of the simulation environment and the next scheduling operation output by the dispatcher's decision model of the quantum business system, and feedback and update the simulation environment;

Set a time period, repeat the above steps to form the quantum business system scheduler's recommended operation sequence, and use the recommended operation sequence to assist the quantum business system scheduler in making decisions. Updating the actual operating data of the current quantum business system to the simulation environment in real time can make the simulation environment consistent with the real quantum business system, and simulate some tasks that the dispatcher needs to perform in the simulation environment according to the trained decision-making model. The list of dispatching operations can be used as a reference for the real dispatcher when performing dispatching operations, so as to prevent the dispatcher from making mistakes due to personal reasons.

In a preferred embodiment of the present invention, using the recommended operation sequence to assist the quantum business system dispatcher to perform decision-making includes:

The dispatcher of the quantum business system refers to the recommended operation sequence and performs scheduling operations in combination with the actual situation;

After the operation is scheduled, the actual operation results are obtained, and the actual operation data of the current quantum business system is updated, and updated in the simulation environment in real time. Update the actual operation results of the dispatcher to the simulation environment, so that the decision-making model can make the most correct decision under the latest data.

In a preferred embodiment of the present invention, also include:

Optimize the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher. Using the simulation environment of the quantum business system, aiming at the actual operation of each quantum business system dispatcher, the reinforcement learning method is used to train the personalized decision-making model of the quantum business system dispatcher, so as to strengthen the individualization of the quantum business system dispatcher Learning the decision-making model of the quantum business system dispatcher As a simulated dispatcher, continuously interact with the simulation environment of the quantum business system, simulate the operation and scheduling of the quantum business system, and evaluate the personalized reinforcement learning of the quantum business system dispatcher’s decision-making model for the quantum business system dispatcher The scheduling results are compared with the optimal scheduling instruction execution strategy to find out the scheduling problems, and then improve the scheduling strategy. According to the actual situation, in the simulation environment of the quantum business system, simulate the scheduling operations of all dispatchers, and compare with the optimal scheduling Comparing instruction execution strategies, discovering abnormal links, optimizing scheduling methods, and according to the actual situation in the future, in the quantum business system simulation environment, combining the actual data of the quantum business system to simulate the operation and scheduling of the quantum business system in the future, so as to detect problems in advance and avoid accidents , continuously collect data from the actual quantum business system operation and scheduling, and use it to optimize the decision-making model of the quantum business system scheduler for reinforcement learning.

In a preferred embodiment of the present invention, optimizing the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher includes:

Obtain the actual operation data of each quantum business system dispatcher;

Use the actual operation data to train the decision-making model of the quantum business system dispatcher in the simulation environment;

Use the quantum business system dispatcher decision model to simulate the operation and scheduling of the quantum business system in a simulation environment;

Compare the scheduling results of the quantum business system dispatcher decision-making model with the optimal scheduling instructions;

According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized.

In a preferred embodiment of the present invention, also include:

Compare the dispatching operation of the dispatcher decision model of the quantum business system with the optimal dispatching instruction;

According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized.

In a preferred embodiment of the present invention, collecting the data of each quantum system unit through the cloud data center includes:

The data of superconducting quantum system, nuclear spin quantum system, optical cavity quantum system and ion well quantum system are collected through the cloud data center.

In a preferred embodiment of the present invention, the cloud data center is converted from the actual quantum business system of each quantum system unit to the business that the cloud center can handle through the quantum business adapter. The quantum business cloud data center is converted from the actual quantum business system of each quantum node to the classic business that the cloud center can handle through the quantum business adapter, and the process creation of the classic business is realized in the cloud center through process arrangement, and provides quantum business Microservice management method adapted to classic business.

In a preferred embodiment of the present invention, also include:

Create microservices in the cloud data center to encapsulate different quantum business processing classes in the cloud data center adaptation interface.

In a preferred embodiment of the present invention, creating microservices in the cloud data center includes:

Obtain a microservice creation request, where the microservice creation request includes creation information and target quantum business;

Find the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business;

Use the target processing class to deploy the creation information to create an instance of the microservice on the target quantum business.

In a preferred embodiment of the present invention, also include:

In response to receiving the microservice creation request input by the user in the cloud data center, obtain the creation information of the creation request and the target quantum business;

Based on the target quantum business, call the corresponding processing class in the cloud data center to establish a microservice instance. It can reduce the difficulty of micro-service application deployment and management, and realize the adaptive management of different types of quantum services.

In a preferred embodiment of the present invention, creating microservices in the cloud data center includes:

Obtain metadata and service type instances corresponding to various types of quantum computers;

orchestrating service type instances to build a software environment for service type instances;

Configure the parameters of the corresponding service type instance according to the metadata to obtain the processing class of each quantum business.

In a preferred embodiment of the present invention, searching for the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business includes:

Judging whether the target quantum business is recorded in the corresponding relationship between the processing class and the quantum business;

In response to recording the target quantum service, use the processing class corresponding to the target quantum service as the target processing class;

In response to no target quantum service being recorded, a pre-set general processing class is invoked as the target processing class.

The invention collects quantum business system data through a large number of sensing devices, uses digital twin technology to form a simulation environment based on massive data, designs a decision-making model for the dispatcher of the quantum business system through reinforcement learning, and adopts A3C training according to the actual situation of the quantum business system dispatcher. The method interacts with the simulation environment, and finally forms the best execution strategy, which is used to assist the quantum business system system dispatcher in decision-making execution, eliminate the impact of human error as much as possible, and improve scheduling efficiency and accuracy.

The present invention encapsulates different quantum business processing classes in the data platform adaptation interface, shielding different deployment and management methods at the bottom of the platform. When a user needs to use a certain type of quantum business to provide services, he can directly input a microservice creation request on the platform management interface, and the management platform can establish a microservice instance on the target quantum business by searching and calling the corresponding processing class, reducing the It eliminates the difficulty of microservice application deployment and management, and realizes the adaptive management of different types of quantum services.

It should be noted that those skilled in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented through computer programs to instruct relevant hardware to complete, and the above programs can be stored in computer-readable storage media. When the program is executed, it may include the processes of the embodiments of the above-mentioned methods. The storage medium can be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random Access Memory, RAM). The foregoing computer program embodiments can achieve the same or similar effects as any of the foregoing method embodiments corresponding thereto.

In addition, the method disclosed according to the embodiments of the present invention can also be implemented as a computer program executed by the CPU, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the CPU, the above-mentioned functions defined in the methods disclosed in the embodiments of the present invention are executed.

Based on the above purpose, the second aspect of the embodiment of the present invention proposes a multi-quantum service conversion and simulation scheduling device, as shown in Figure 2, the device 200 includes:

A building module, the building module is configured to collect the data of each quantum system unit through the cloud data center, and build a quantum business system simulation environment based on the collected data;

A design module, the design module is configured to design a quantum business system dispatcher decision model;

A training module, the training module is configured to obtain the data of the decision-making execution of the quantum business system dispatcher, and train the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the obtained data of the execution decision-making;

An execution module configured to use the trained quantum business system dispatcher decision model to assist the quantum business system dispatcher to execute decisions.

Based on the above purpose, a third aspect of the embodiments of the present invention provides a computer device. FIG. 3 shows a schematic diagram of an embodiment of a computer device provided by the present invention. As shown in FIG. 3 , the embodiment of the present invention includes the following devices: at least one processor 21; and a memory 22, the memory 22 stores computer instructions 23 that can run on the processor, and when the instructions are executed by the processor, the following methods are implemented:

Collect the data of each quantum system unit through the cloud data center, and build a quantum business system simulation environment based on the collected data;

Design the decision-making model of the dispatcher of the quantum business system;

Obtain the decision-making data of the dispatcher of the quantum business system, and train the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the obtained data of the decision-making execution;

Use the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher in making decisions.

In a preferred embodiment of the present invention, collecting the data of each quantum system unit through the cloud data center, and constructing a quantum business system simulation environment based on the collected data includes:

Collect the operation data of each quantum system unit, wherein the operation data includes the real-time status data, operation log, scheduling plan, fault data and dispatcher scheduling execution data of each quantum system unit;

Based on the collected data and using digital twin technology to build a quantum business system simulation environment.

In a preferred embodiment of the present invention, obtaining the data of the execution decision-making of the quantum business system dispatcher, and training the decision-making model of the quantum business system dispatcher in a simulation environment based on the obtained data of execution decision-making includes:

Obtain the data of the decision-making of the dispatcher of the quantum business system;

According to the actual operation of the quantum business system dispatcher, combined with the state environment of the quantum business system, the quantum business system dispatcher dispatches the operation instruction execution sequence;

According to the current quantum business system plan, the operating status of each quantum system unit and the practice of abnormal accidents, determine the next step to execute the scheduling operation in the simulation environment;

According to the scheduling operation of the dispatcher of the quantum business system and the actual execution effect, combined with the scheduling operation determined in the simulation environment, the reward function of the simulation environment of the quantum business system is set;

Based on the reward function and using the A3C algorithm to train the decision-making model of the quantum business system dispatcher.

In a preferred embodiment of the present invention, based on the reward function and adopting the A3C algorithm to train the quantum business system dispatcher decision-making model includes:

Set the number of worker threads, the number of global shared iterations, the global maximum number of iterations, the state feature dimension and the global parameters of the operation instruction set, set the global model public neural network, and set the initialization state of the simulation environment;

Initialize the decision-making model of the quantum business system dispatcher;

Make each worker thread use the global model public neural network to independently interact with the simulation environment, perform scheduling operations to obtain feedback and update the gradient of the local global model public neural network, and update the model parameters of the global model public neural network;

Repeat the previous step until the quantum business system dispatcher decision model converges.

In a preferred embodiment of the present invention, the global model public neural network includes an Actor network and a Critic network.

In a preferred embodiment of the present invention, using the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher to perform decision-making includes:

Collect the actual operating data of the current quantum business system in real time and update it to the simulation environment in real time;

Use the trained quantum business system dispatcher decision-making model to output the next step of dispatching operation according to the current actual situation;

Record the current state of the simulation environment and the next scheduling operation output by the dispatcher's decision model of the quantum business system, and feedback and update the simulation environment;

Set a time period, repeat the above steps to form the quantum business system scheduler's recommended operation sequence, and use the recommended operation sequence to assist the quantum business system scheduler in making decisions.

In a preferred embodiment of the present invention, using the recommended operation sequence to assist the quantum business system dispatcher to perform decision-making includes:

The dispatcher of the quantum business system refers to the recommended operation sequence and performs scheduling operations in combination with the actual situation;

After the operation is scheduled, the actual operation results are obtained, and the actual operation data of the current quantum business system is updated, and updated in the simulation environment in real time.

In a preferred embodiment of the present invention, also include:

Optimize the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher.

In a preferred embodiment of the present invention, optimizing the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher includes:

Obtain the actual operation data of each quantum business system dispatcher;

Use the actual operation data to train the decision-making model of the quantum business system dispatcher in the simulation environment;

Use the quantum business system dispatcher decision model to simulate the operation and scheduling of the quantum business system in a simulation environment;

Compare the scheduling results of the quantum business system dispatcher decision-making model with the optimal scheduling instructions;

According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized.

In a preferred embodiment of the present invention, also include:

Compare the dispatching operation of the dispatcher decision model of the quantum business system with the optimal dispatching instruction;

According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized.

In a preferred embodiment of the present invention, collecting the data of each quantum system unit through the cloud data center includes:

The data of superconducting quantum system, nuclear spin quantum system, optical cavity quantum system and ion well quantum system are collected through the cloud data center.

In a preferred embodiment of the present invention, the cloud data center is converted from the actual quantum business system of each quantum system unit to the business that the cloud center can handle through the quantum business adapter.

In a preferred embodiment of the present invention, also include:

Create microservices in the cloud data center to encapsulate different quantum business processing classes in the cloud data center adaptation interface.

In a preferred embodiment of the present invention, creating microservices in the cloud data center includes:

Obtain a microservice creation request, where the microservice creation request includes creation information and target quantum business;

Find the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business;

Use the target processing class to deploy the creation information to create an instance of the microservice on the target quantum business.

In a preferred embodiment of the present invention, also include:

In response to receiving the microservice creation request input by the user in the cloud data center, obtain the creation information of the creation request and the target quantum business;

Based on the target quantum business, call the corresponding processing class in the cloud data center to establish a microservice instance.

In a preferred embodiment of the present invention, creating microservices in the cloud data center includes:

Obtain metadata and service type instances corresponding to various types of quantum computers;

orchestrating service type instances to build a software environment for service type instances;

Configure the parameters of the corresponding service type instance according to the metadata to obtain the processing class of each quantum business.

In a preferred embodiment of the present invention, searching for the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business includes:

Judging whether the target quantum business is recorded in the corresponding relationship between the processing class and the quantum business;

In response to recording the target quantum service, use the processing class corresponding to the target quantum service as the target processing class;

In response to no target quantum service being recorded, a pre-set general processing class is invoked as the target processing class.

Based on the above purpose, a fourth aspect of the embodiments of the present invention provides a computer-readable storage medium. FIG. 4 is a schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. As shown in FIG. 4 , a computer readable storage medium 31 stores a computer program 32 for executing the above method when executed by a processor.

In addition, the method disclosed according to the embodiments of the present invention can also be implemented as a computer program executed by a processor, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the above functions defined in the methods disclosed in the embodiments of the present invention are executed.

In addition, the above-mentioned method steps and system units can also be realized by using a controller and a computer-readable storage medium for storing a computer program for enabling the controller to realize the functions of the above-mentioned steps or units.

Those of skill would also appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described generally in terms of their functionality. Whether such functionality is implemented as software or as hardware depends upon the particular application and design constraints imposed on the overall system. Those skilled in the art may implement the functions in various ways for each specific application, but such implementation decisions should not be interpreted as causing a departure from the scope disclosed in the embodiments of the present invention.

In one or more exemplary designs, functions may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. Storage media may be any available media that can be accessed by a general purpose or special purpose computer. By way of example and not limitation, the computer readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage device, magnetic disk storage device or other magnetic storage device, or may be used to carry or store instructions in Any other medium that can be accessed by a general purpose or special purpose computer or a general purpose or special purpose processor, and the required program code or data structure. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial Cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of media. Disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers . Combinations of the above should also be included within the scope of computer-readable media.

The above are the exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications can be made without departing from the scope of the disclosed embodiments of the present invention defined in the claims. The functions, steps and/or actions of the method claims in accordance with the disclosed embodiments described herein need not be performed in any particular order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in an individual form, they may also be understood as a plurality unless explicitly limited to a singular number.

It should be understood that as used herein, the singular form "a" and "an" are intended to include the plural forms as well, unless the context clearly supports an exception. It should also be understood that "and/or" as used herein is meant to include any and all possible combinations of one or more of the associated listed items.

The serial numbers of the embodiments disclosed in the above-mentioned embodiments of the present invention are only for description, and do not represent the advantages and disadvantages of the embodiments.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above-mentioned embodiments can be completed by hardware, or can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned The storage medium may be a read-only memory, a magnetic disk or an optical disk, and the like.

Those of ordinary skill in the art should understand that: the discussion of any of the above embodiments is exemplary only, and is not intended to imply that the disclosed scope (including claims) of the embodiments of the present invention is limited to these examples; under the idea of the embodiments of the present invention , the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes in different aspects of the above embodiments of the present invention, which are not provided in details for the sake of brevity. Therefore, within the spirit and principle of the embodiments of the present invention, any omissions, modifications, equivalent replacements, improvements, etc., shall be included in the protection scope of the embodiments of the present invention.

Claims (20)

1. A method for multi-quantum business conversion and simulation scheduling, characterized in that, comprising the following steps: Collect the data of each quantum system unit through the cloud data center, and build a quantum business system simulation environment based on the collected data; Design the decision-making model of the dispatcher of the quantum business system; Obtain the decision-making data of the dispatcher of the quantum business system, and train the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the obtained data of the decision-making execution; Use the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher in making decisions. 2. The method according to claim 1, characterized in that, collecting the data of each quantum system unit through the cloud data center, and constructing a quantum business system simulation environment based on the collected data includes: Collect the operation data of each quantum system unit, wherein the operation data includes the real-time status data, operation log, scheduling plan, fault data and dispatcher scheduling execution data of each quantum system unit; Based on the collected data and using digital twin technology to build a quantum business system simulation environment. 3. The method according to claim 1, characterized in that, obtaining the data of the execution decision of the quantum business system dispatcher, and training the decision model of the quantum business system dispatcher in the simulated simulation environment based on the data of the execution decision obtained includes: Obtain the data of the decision-making of the dispatcher of the quantum business system; According to the actual operation of the quantum business system dispatcher, combined with the state environment of the quantum business system, the quantum business system dispatcher dispatches the operation instruction execution sequence; According to the current quantum business system plan, the operating status of each quantum system unit and the practice of abnormal accidents, determine the next step to execute the scheduling operation in the simulation environment; According to the scheduling operation of the dispatcher of the quantum business system and the actual execution effect, combined with the scheduling operation determined in the simulation environment, the reward function of the simulation environment of the quantum business system is set; Based on the reward function and using the A3C algorithm to train the decision-making model of the quantum business system dispatcher. 4. method according to claim 3, is characterized in that, based on reward function and adopts A3C algorithm to train quantum service system dispatcher's decision-making model to comprise: Set the number of worker threads, the number of global shared iterations, the global maximum number of iterations, the state feature dimension and the global parameters of the operation instruction set, set the global model public neural network, and set the initialization state of the simulation environment; Initialize the decision-making model of the quantum business system dispatcher; Make each worker thread use the global model public neural network to independently interact with the simulation environment, perform scheduling operations to obtain feedback and update the gradient of the local global model public neural network, and update the model parameters of the global model public neural network; Repeat the previous step until the quantum business system dispatcher decision model converges. 5. method according to claim 4, is characterized in that, global model public neural network comprises Actor network and Critic network. 6. The method according to claim 1, wherein, using the trained quantum business system dispatcher decision-making model to assist the quantum business system dispatcher to perform decision-making includes: Collect the actual operating data of the current quantum business system in real time and update it to the simulation environment in real time; Use the trained quantum business system dispatcher decision-making model to output the next step of dispatching operation according to the current actual situation; Record the current state of the simulation environment and the next scheduling operation output by the dispatcher's decision model of the quantum business system, and feedback and update the simulation environment; Set a time period, repeat the above steps to form the quantum business system scheduler's recommended operation sequence, and use the recommended operation sequence to assist the quantum business system scheduler in making decisions. 7. The method according to claim 6, characterized in that, using the recommended operation sequence to assist the quantum business system dispatcher to execute the decision comprises: The dispatcher of the quantum business system refers to the recommended operation sequence and performs scheduling operations in combination with the actual situation; After the operation is scheduled, the actual operation results are obtained, and the actual operation data of the current quantum business system is updated, and updated in the simulation environment in real time. 8. The method of claim 1, further comprising: Optimize the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher. 9. The method according to claim 8, characterized in that, optimizing the decision-making model of the quantum business system dispatcher according to the actual operation of each quantum business system dispatcher comprises: Obtain the actual operation data of each quantum business system dispatcher; Use the actual operation data to train the decision-making model of the quantum business system dispatcher in the simulation environment; Use the quantum business system dispatcher decision model to simulate the operation and scheduling of the quantum business system in a simulation environment; Compare the scheduling results of the quantum business system dispatcher decision-making model with the optimal scheduling instructions; According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized. 10. The method of claim 9, further comprising: Compare the dispatching operation of the dispatcher decision model of the quantum business system with the optimal dispatching instruction; According to the comparison results, the dispatching strategy of the dispatcher decision model of the quantum business system is optimized. 11. The method according to claim 1, wherein collecting the data of each quantum system unit through the cloud data center comprises: The data of superconducting quantum system, nuclear spin quantum system, optical cavity quantum system and ion well quantum system are collected through the cloud data center. 12. The method according to claim 1, wherein the cloud data center is converted from the actual quantum business system of each quantum system unit into a business that can be processed by the cloud center through a quantum business adapter. 13. The method of claim 1, further comprising: Create microservices in the cloud data center to encapsulate different quantum business processing classes in the cloud data center adaptation interface. 14. The method according to claim 13, wherein creating microservices in the cloud data center comprises: Obtain a microservice creation request, where the microservice creation request includes creation information and target quantum business; Find the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business; Use the target processing class to deploy the creation information to create an instance of the microservice on the target quantum business. 15. The method of claim 14, further comprising: In response to receiving the microservice creation request input by the user in the cloud data center, obtain the creation information of the creation request and the target quantum business; Based on the target quantum business, call the corresponding processing class in the cloud data center to establish a microservice instance. 16. The method according to claim 14, wherein creating microservices in the cloud data center comprises: Obtain metadata and service type instances corresponding to various types of quantum computers; orchestrating service type instances to build a software environment for service type instances; Configure the parameters of the corresponding service type instance according to the metadata to obtain the processing class of each quantum business. 17. The method according to claim 16, wherein searching for the target processing class corresponding to the target quantum business according to the corresponding relationship between the processing class and the quantum business includes: Judging whether the target quantum business is recorded in the corresponding relationship between the processing class and the quantum business; In response to recording the target quantum service, use the processing class corresponding to the target quantum service as the target processing class; In response to no target quantum service being recorded, a pre-set general processing class is invoked as the target processing class. 18. A device for multi-quantum service conversion and simulation scheduling, characterized in that the device includes: A building module, the building module is configured to collect the data of each quantum system unit through the cloud data center, and build a quantum business system simulation environment based on the collected data; A design module, the design module is configured to design a quantum business system dispatcher decision model; A training module, the training module is configured to obtain the data of the decision-making execution of the quantum business system dispatcher, and train the decision-making model of the dispatcher of the quantum business system in a simulation environment based on the obtained data of the execution decision-making; An execution module configured to use the trained quantum business system dispatcher decision model to assist the quantum business system dispatcher to execute decisions. 19. A computer device, comprising: at least one processor; and A memory, the memory stores computer instructions operable on the processor, and when the instructions are executed by the processor, the steps of the method according to any one of claims 1-17 are implemented. 20. A computer-readable storage medium, the computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1-17 are implemented.

Images