Fusion computing system and task processing method

Abstract

The invention discloses a fusion computing system and a task processing method. The system comprises a task processing server, a message proxy server and quantum computing power resource equipment. The task processing server can process the obtained calculation task to obtain a quantum calculation sub-task and send the quantum calculation sub-task to the message proxy server. And the message proxy server can store the quantum computing subtask sent by the task processing server based on the first message queue. The quantum computing power resource device can process the quantum computing subtask acquired from the message proxy server to obtain a target quantum processing result, and returns the target quantum processing result to the message proxy server. And the message proxy server can store the target quantum processing result returned by the quantum computing power resource device based on the second message queue. And the task processing server can obtain the target quantum processing result from the message proxy server. Therefore, asynchronism and decoupling are realized, and the expandability of the system is enhanced.

Description

Technical Field

[0001] The present application relates to the field of quantum computing, and more specifically, to a fusion computing system and task processing method. Background Art

[0002] Superconducting quantum computing devices that use superconducting circuits to implement quantum bits can use macroscopic quantum states to encode information, generate strong coupling with control and reading circuits, and achieve high-speed quantum information processing. They have significant advantages in solving specific problems such as factoring large numbers and searching for unstructured data. However, due to the limitations of existing hardware technology and interference from the external environment, superconducting quantum computing devices still face many challenges in practical applications. Summary of the invention

[0003] The present application provides a fusion computing system and a task processing method.

[0004] The embodiment of the present application provides a fusion computing system, which includes a task processing server, a message proxy server and a quantum computing power resource device;

[0005] The task processing server is configured to process the acquired computing task to obtain a quantum computing subtask, and send the quantum computing subtask to the message proxy server;

[0006] The message proxy server includes a first message queue associated with the quantum computing power resource device, and the message proxy server is configured to store the quantum computing subtask sent by the task processing server based on the first message queue;

[0007] The quantum computing power resource device is configured to process the quantum computing subtask obtained from the message proxy server, obtain a target quantum processing result, and return it to the message proxy server;

[0008] The message proxy server includes a second message queue associated with the quantum computing power resource device, and the message proxy server is configured to store the target quantum processing result returned by the quantum computing power resource device based on the second message queue;

[0009] The task processing server is configured to obtain the target quantum processing result from the message proxy server.

[0010] Thus, the present application provides a fusion computing system, which includes a message proxy server, a task processing server and a quantum computing power resource device. First, the task processing server can process the acquired computing task to obtain a quantum computing subtask, and send the quantum computing subtask to the message proxy server. Then, the message proxy server includes a first message queue associated with the quantum computing power resource device, and based on the first message queue, the message proxy server can store the quantum computing subtask sent by the task processing server. The quantum computing power resource device then processes the quantum computing subtask obtained from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server. Then, the message proxy server includes a second message queue associated with the quantum computing power resource device, and based on the second message queue, the message proxy server can store the target quantum processing result returned by the quantum computing power resource device. Finally, the task processing server is configured to obtain the target quantum processing result from the message proxy server. In this way, the fusion computing system realizes the effective fusion of classical computing and quantum computing resources, and improves the execution efficiency of computing tasks. At the same time, based on the message queue in the message proxy server, the fusion computing system realizes the asynchrony and decoupling of task processing, and enhances the scalability and robustness of the system.

[0011] In some embodiments, the task processing server is configured to:

[0012] In response to a resource selection operation, a target quantum computing device is determined from the quantum computing power resource device.

[0013] In this way, the task processing server can determine the target quantum computing device from the quantum computing power resource device in response to the resource selection operation. In this way, the user can select the appropriate quantum computing device according to the actual needs, and the task processing server can determine the target quantum computing device in response to the resource selection operation, and process the quantum computing subtask according to the target quantum computing device in the subsequent process, thereby optimizing the entire computing task processing process and improving the task processing efficiency and task success rate.

[0014] In some embodiments, the task processing server is configured to:

[0015] Obtaining a first subscription identifier associated with the target quantum computing device;

[0016] subscribing to a first target message queue in the message proxy server according to the first subscription identifier, where the first target message queue is used to indicate a message queue associated with the target quantum computing device among a plurality of first message queues of the message proxy server;

[0017] Sending the quantum computing subtask to the first target message queue.

[0018] In this way, the task processing server can obtain the first subscription identifier associated with the target quantum computing device. Then, the task processing server can subscribe to the first target message queue in the message proxy server according to the first subscription identifier, and the first target message queue is used to indicate the message queue associated with the target quantum computing device in multiple first message queues of the message proxy server. Then, the task processing server can send the quantum computing subtask to the first target message queue. In this way, through the first subscription identifier and the first target message queue, the task processing server can effectively send the quantum computing subtask to the target quantum computing device and ensure that the quantum computing subtask can be correctly received and processed. At the same time, this communication method based on message queues also provides asynchronous processing capabilities, so that the task processing server can continue to perform other operations after sending the quantum computing subtask without waiting for the quantum computing subtask to be processed, thereby improving the overall efficiency of the system.

[0019] In some embodiments, the quantum computing power resource device is configured to:

[0020] Obtaining the first subscription identifier;

[0021] Subscribing to the first target message queue according to the first subscription identifier;

[0022] Obtain the quantum computing subtask from the first target message queue;

[0023] The quantum computing subtask is processed to obtain an initial quantum processing result.

[0024] In this way, the quantum computing power resource device can obtain the first subscription identifier. Next, the quantum computing power resource device can subscribe to the first target message queue according to the first subscription identifier. Then, the quantum computing power resource device can obtain the quantum computing subtask from the first target message queue. Finally, the quantum computing power resource device can process the quantum computing subtask to obtain the initial quantum processing result. In this way, through the first subscription identifier and the first target message queue, the quantum computing power resource device can effectively obtain the quantum computing subtask to be processed from the message proxy server, and perform corresponding processing to obtain the initial quantum processing result. This communication method based on message queues also provides asynchronous processing capabilities, so that the quantum computing power resource device can continue to listen to the message queue and obtain new tasks while processing quantum computing subtasks.

[0025] In some embodiments, the quantum computing power resource device is configured to:

[0026] Based on a preset security mechanism, the quantum computing subtask is verified to confirm the validity of the source of the quantum computing subtask.

[0027] In this way, based on the preset security mechanism, the quantum computing resource device can verify the quantum computing subtask to confirm the validity of the source of the quantum computing subtask. In this way, based on the preset security mechanism, the quantum computing resource device can ensure that the source of the received quantum computing subtask is valid, the sender is legal, the quantum computing subtask remains intact during the transmission process, and the quantum computing subtask is within the validity period. This helps protect the quantum computing resource device from malicious attacks and illegal access, and ensures the security and reliability of quantum computing.

[0028] In some embodiments, the quantum computing power resource device is configured to:

[0029] Obtaining a second subscription identifier associated with the target quantum computing device;

[0030] Subscribe to a second target message queue according to the second subscription identifier, where the second target message queue is used to indicate a message queue associated with the target quantum computing device among a plurality of second message queues of the message proxy server;

[0031] The initial quantum processing result is optimized based on a preset optimization method to obtain the target quantum processing result, and the target quantum processing result is sent to the second target message queue.

[0032] In this way, the quantum computing power resource device can obtain the second subscription identifier associated with the target quantum computing device. Then, the quantum computing power resource device can subscribe to the second target message queue according to the second subscription identifier, and the second target message queue is used to indicate the message queue associated with the target quantum computing device in multiple second message queues of the message proxy server. Finally, the quantum computing power resource device can optimize the initial quantum processing result based on a preset optimization method, obtain the target quantum processing result, and send the target quantum processing result to the second target message queue. In this way, through the communication method based on the message queue, the ability of asynchronous processing can be provided, so that the quantum computing power resource device can continue to monitor the message queue and obtain new tasks while processing tasks.

[0033] In some embodiments, the task processing server is configured to:

[0034] Obtaining the second subscription identifier;

[0035] Subscribing to the second target message queue in the message proxy server according to the second subscription identifier;

[0036] The target quantum processing result is obtained from the second target message queue.

[0037] In this way, the task processing server can obtain the second subscription identifier. Then, the task processing server can subscribe to the second target message queue in the message proxy server according to the second subscription identifier. Finally, the task processing server can obtain the target quantum processing result from the second target message queue. In this way, the communication method based on the message queue can provide asynchronous processing capabilities, so that the task processing server can continue to perform other operations after sending the task without waiting for the task processing to be completed. At the same time, this decoupled design also enhances the scalability and robustness of the system, allowing the task processing server to easily interact with multiple quantum computing power resource devices.

[0038] In certain embodiments, the fusion computing system further includes a classic resource access device and a classic computing power resource device;

[0039] The task processing server is configured to process the computing task to obtain a classic computing subtask;

[0040] The classic resource access device is configured to call the classic computing resource device to perform computational processing on the classic computing subtask to obtain a target classic processing result.

[0041] In this way, the fusion computing system also includes a classical resource access device and a classical computing power resource device. The task processing server can process the computing task to obtain a classical computing subtask. The classical resource access device can call the classical computing power resource device to perform computing processing on the classical computing subtask to obtain the target classical processing result. In this way, the fusion computing system can make full use of classical computing resources and quantum computing resources to achieve efficient processing of computing tasks, and the fusion computing system can flexibly process various types of computing tasks, thereby improving the overall computing power and efficiency.

[0042] The embodiment of the present application provides a task processing method, which is based on the above-mentioned fusion computing system and includes:

[0043] The task processing server processes the acquired computing task to obtain a quantum computing subtask, and sends the quantum computing subtask to the message proxy server;

[0044] The message proxy server stores the quantum computing subtask sent by the task processing server based on the first message queue;

[0045] The quantum computing power resource device processes the quantum computing subtask obtained from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server;

[0046] The message proxy server stores the target quantum processing result returned by the quantum computing power resource device based on the second message queue;

[0047] The task processing server obtains the target quantum processing result from the message proxy server.

[0048] In this way, the task processing server processes the acquired computing task to obtain a quantum computing subtask, and sends the quantum computing subtask to the message proxy server. Then, the message proxy server stores the quantum computing subtask sent by the task processing server based on the first message queue. The quantum computing power resource device then processes the quantum computing subtask obtained from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server. Then, based on the second message queue, the message proxy server stores the target quantum processing result returned by the quantum computing power resource device. Finally, the task processing server obtains the target quantum processing result from the message proxy server. In this way, the fusion computing system can make full use of quantum computing resources to achieve efficient processing of computing tasks. The message proxy server can store and forward quantum computing subtasks and quantum processing results to ensure effective communication between the task processing server and the quantum computing power resource device. This communication method based on message queues can also provide asynchronous processing capabilities, so that the task processing server can continue to perform other operations after sending the task without waiting for the task processing to complete. At the same time, through this decoupled design, the scalability and robustness of the system are enhanced, so that the task processing server can easily interact with multiple quantum computing power resource devices.

[0049] In certain embodiments, the method further comprises:

[0050] The task processing server processes the computing task to obtain a classic computing subtask;

[0051] The classic resource access device calls the classic computing resource device to perform computational processing on the classic computing subtask to obtain a target classic processing result.

[0052] In this way, the task processing server processes the computing task to obtain the classic computing subtask. Then, the classic resource access device calls the classic computing resource device to perform computing processing on the classic computing subtask to obtain the target classic processing result. In this way, the fusion computing system can make full use of the classic computing resources to achieve efficient processing of computing tasks. Based on the classic resource access device and the classic computing resource device, the system can flexibly handle various types of computing tasks, improving the overall computing power and efficiency. At the same time, this message queue-based communication method can also provide asynchronous processing capabilities, allowing the task processing server to continue to perform other operations after sending the task without waiting for the task processing to complete.

[0053] Additional aspects and advantages of the embodiments of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0055] Figure 1 is a schematic diagram of the structure of a fusion computing system of certain embodiments of the present application;

[0056] Figure 2 is a schematic diagram of a quantum computing subtask processing flow of certain embodiments of the present application;

[0057] Figure 3 It is a schematic diagram of the processing flow of the classical computing subtasks of certain implementation modes of the present application;

[0058] Figure 4 It is one of the flowcharts of the task processing method of certain implementation modes of the present application;

[0059] Figure 5 This is the second flowchart of the task processing method of certain implementation modes of the present application;

[0060] Figure 6 is a schematic diagram of a hybrid neural network model of certain embodiments of the present application;

[0061] Figure 7 is a schematic diagram of a quantum computing cloud platform of certain embodiments of the present application. DETAILED DESCRIPTION

[0062] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of the present application, and cannot be understood as limiting the embodiments of the present application.

[0063] Superconducting quantum computing devices, which use superconducting circuits to realize quantum bits, are a highly promising quantum computing solution. This device can use macroscopic quantum states to encode information and generate strong coupling with control and reading circuits to achieve high-speed quantum information processing. Superconducting quantum computing devices have significant advantages in solving specific problems such as factoring large numbers and searching unstructured data.

[0064] Factoring large numbers is an important problem in cryptography and is of great significance for protecting information security. When dealing with the factorization of large numbers, the difficulty of traditional calculation methods increases exponentially as the number of digits increases. Superconducting quantum computing devices use the superposition and entanglement characteristics of quantum bits to complete the factorization of large numbers in a short period of time, thereby cracking existing encryption algorithms and posing a threat to information security.

[0065] Unstructured data search refers to the process of finding specific information in a large amount of disordered data. Traditional computing methods need to traverse all data when processing unstructured data search, which is inefficient. Superconducting quantum computing devices use the superposition and entanglement characteristics of quantum bits to complete the search of unstructured data in a short time, thereby improving search efficiency and providing a new solution for big data processing.

[0066] However, due to the limitations of existing hardware technology and interference from the external environment, superconducting quantum computing devices still face many challenges in practical applications. First, the preparation and manipulation of superconducting quantum bits require extremely high precision and stability. Secondly, superconducting quantum computing devices need to operate at extremely low temperatures and have extremely harsh environmental requirements, which limits the practicality and scalability of the device. In addition, superconducting quantum computing devices have a high error rate, and further technological innovation and optimization are needed to achieve reliable quantum computing.

[0067] Based on the above questions, please refer to Figure 1 An embodiment of the present application provides a fusion computing system, which includes a task processing server, a message proxy server and a quantum computing power resource device.

[0068] The task processing server is configured to process the acquired computing task to obtain a quantum computing subtask, and send the quantum computing subtask to the message proxy server;

[0069] The message proxy server includes a first message queue associated with the quantum computing power resource device, and the message proxy server is configured to store the quantum computing subtask sent by the task processing server based on the first message queue;

[0070] The quantum computing power resource device is configured to process the quantum computing subtasks obtained from the message proxy server, obtain the target quantum processing results, and return them to the message proxy server;

[0071] The message proxy server includes a second message queue associated with the quantum computing power resource device, and the message proxy server is configured to store a target quantum processing result returned by the quantum computing power resource device based on the second message queue;

[0072] The task processing server is configured to obtain the target quantum processing result from the message proxy server.

[0073] Specifically, the fusion computing system is a hybrid quantum computing structure, which is a new computing architecture that combines the parallel processing capabilities of quantum computing resources with the efficient numerical computing capabilities of supercomputers, aiming to enhance the flexibility and efficiency of algorithms, improve computing speed, optimize resource allocation, and expand the scope of application. By integrating quantum computing resources with supercomputers, it is possible to not only take advantage of the unique advantages of quantum computing, but also give play to the stability and reliability of supercomputing, achieve the complementary advantages of the two, and then solve complex problems and improve the efficiency of solving complex problems.

[0074] The task processing server can receive computing tasks, process them to generate quantum computing subtasks, and send these quantum computing subtasks to the first message queue of the message proxy server. In addition, the task processing server is also responsible for obtaining the target quantum processing results returned by the quantum computing resource device from the second message queue of the message proxy server.

[0075] A computing task refers to a series of operations or algorithms that need to be performed by a computing device to solve a specific problem or complete a task.

[0076] Quantum computing subtasks involve specific quantum algorithms and operations that exploit the special properties of quantum bits, such as superposition and entanglement, to perform calculations.

[0077] Quantum circuit is a concept in quantum computing that is used to describe the order of operations between quantum bits and the way quantum gates are connected. A quantum circuit consists of a series of quantum gates that act on quantum bits in a certain order to implement specific quantum algorithms or operations.

[0078] Quantum gate is the basic operation unit of quantum computing, which can transform the state of quantum bits and realize the processing of quantum information. The function of quantum gate is to create entanglement between quantum bits or change the superposition state of quantum bits.

[0079] The message proxy server includes multiple message queues associated with different quantum computing resource devices. The message proxy server is equivalent to an intermediary component that can receive and store data information sent by the task processing server and the quantum computing resource device. Through the message proxy server, the fusion computing system can shield the hardware technical details of the computing resource, allowing the task processing server to uniformly control and schedule the quantum computing resource device, thus lowering the user's usage threshold.

[0080] The message queue can be regarded as a middleware for asynchronous communication, allowing devices to communicate by sending and receiving messages without knowing the exact location or status of each other. In the implementation of the present application, communication based on the message queue can achieve decoupling between the task processing server and the quantum computing resource device. The two do not need to communicate directly with each other, but interact through the message queue. At the same time, asynchronous communication is allowed between the task processing server and the quantum computing resource device.

[0081] Decoupling refers to reducing the dependencies between different components or modules to a minimum so that they can be developed, tested, and deployed independently. Decoupling between different components or modules can improve the maintainability, scalability, and reusability of the system.

[0082] Asynchronous communication refers to the fact that the sending and receiving of information do not need to occur at the same time, nor does it require a real-time connection between the sender and the receiver. In asynchronous communication, the sender sends the information to an intermediate storage and can then continue to perform other tasks without waiting for an immediate response from the receiver. The receiver can retrieve and process the information from the intermediate storage at any time.

[0083] The first message queue refers to the message queue used to process data information sent by the task processing server to the quantum computing resource device.

[0084] The second message queue refers to the message queue used to process the data information returned by the quantum computing resource device to the task processing server.

[0085] Quantum computing power resource devices refer to quantum hardware or quantum software environments that can process quantum computing subtasks, including a variety of quantum computers and quantum simulators. These quantum computing power resource devices are built based on different technical routes and are managed and controlled by quantum computing operating systems. Quantum computing power resource devices process quantum computing subtasks obtained from the message queue associated with themselves in the message proxy server, generate target quantum processing results, and return these results to the message proxy server for the task processing server to obtain.

[0086] The Quantum Operation System (QOS) is designed for quantum computing resource devices. It can effectively manage and control quantum computing resource devices and provide a convenient programming interface for users to write and execute quantum programs. It is a system software that controls and manages quantum computing resource devices and provides a program development and operation environment.

[0087] A supercomputer is an extremely powerful computer capable of performing large-scale, complex scientific calculations and simulations to solve scientific, engineering, and business problems that require enormous computing power.

[0088] The supercomputing center is a centralized data processing and computing platform that uses high-performance computer clusters and advanced software tools to provide customers with high-speed, high-precision data processing and analysis capabilities. The supercomputing center includes multiple high-performance computer nodes, storage systems and network equipment, with rich storage capacity and high-speed network bandwidth, which can cope with large-scale scientific computing and data processing needs.

[0089] See also Figure 2 , Figure 2 The figure is a schematic diagram of the quantum computing subtask processing flow of the fusion computing system. The task processing server is the top-level module of the entire scheduling system, responsible for coordinating and managing the workflow of each subsystem and component. It ensures the splitting, allocation and execution of computing tasks, and is also responsible for monitoring the operating status of the entire scheduling system to ensure the stability and reliability of the scheduling system.

[0090] The message proxy server is located at the second layer and is used to store and process data streams from different sources to ensure the orderly transmission and reception of data. The role of the data queue is to decouple data producers and consumers, improving the scalability and flexibility of the system. In addition, the message proxy server can also cache data to prevent data loss and ensure data consistency and reliability.

[0091] The quantum computing resource device is the third layer, which can realize the following functions: computing task source authentication, task parameter verification, and return of computing results. Computing task source authentication refers to the identity authentication and security check of computing tasks to ensure that only authorized users or devices can access and use quantum computing resource devices to process computing tasks. Task parameter verification refers to the validity check of the input task parameters to ensure the correctness and reliability of the computing tasks, so as to avoid task processing failure or erroneous results caused by parameter errors. Running computing tasks refers to executing specific quantum computing tasks and performing data processing and analysis according to the set parameters and algorithms. Returning computing results refers to feeding back the processed experimental results to users or other related systems for further analysis and utilization.

[0092] In summary, the present application provides a fusion computing system, which includes a message proxy server, a task processing server and a quantum computing power resource device. First, the task processing server can process the acquired computing task to obtain a quantum computing subtask, and send the quantum computing subtask to the message proxy server. Then, the message proxy server includes a first message queue associated with the quantum computing power resource device, and based on the first message queue, the message proxy server can store the quantum computing subtask sent by the task processing server. The quantum computing power resource device then processes the quantum computing subtask obtained from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server. Then, the message proxy server includes a second message queue associated with the quantum computing power resource device, and based on the second message queue, the message proxy server can store the target quantum processing result returned by the quantum computing power resource device. Finally, the task processing server is configured to obtain the target quantum processing result from the message proxy server. In this way, the fusion computing system realizes the effective fusion of classical computing and quantum computing resources, and improves the execution efficiency of computing tasks. At the same time, based on the message queue in the message proxy server, the fusion computing system realizes the asynchrony and decoupling of task processing, and enhances the scalability and robustness of the system.

[0093] In some embodiments, the task processing server is configured to:

[0094] In response to the resource selection operation, a target quantum computing device is determined from the quantum computing power resource device.

[0095] Specifically, users can select the most suitable quantum computing device to perform tasks based on the requirements of the task and the current status of the quantum computing resource device. The selection criteria may include the computing power, availability, historical performance data, etc. of the device.

[0096] The task processing server determines the target quantum computing device according to the user's resource selection operation. It should be noted that the target quantum computing device can be multiple quantum computing devices in the quantum computing power resource device, or it can be a quantum computing device in the quantum computing power resource device, which is determined according to actual needs and is not limited here.

[0097] In this way, users can select appropriate quantum computing devices according to actual needs, and the task processing server can determine the target quantum computing device in response to the resource selection operation, and process the quantum computing subtasks according to the target quantum computing device in the subsequent process, thereby optimizing the entire computing task processing process and improving task processing efficiency and task success rate.

[0098] In some embodiments, the task processing server is configured to:

[0099] Obtaining a first subscription identifier associated with a target quantum computing device;

[0100] Subscribe to a first target message queue in the message proxy server according to the first subscription identifier, where the first target message queue is used to indicate a message queue associated with a target quantum computing device in a plurality of first message queues of the message proxy server;

[0101] Send the quantum computing subtask to the first target message queue.

[0102] Specifically, a Subscription Identifier refers to a unique identifier used to identify the subscription relationship of a subscriber (task processing server or target quantum computing device) to a specific message topic. The Subscription Identifier can help track and manage different subscriptions of subscribers to message streams, ensuring that messages can be correctly distributed to the corresponding subscribers.

[0103] The first target message queue refers to a message queue used to process data information sent from the task processing server to the target quantum computing device.

[0104] First, the task processing server needs to obtain a first subscription identifier associated with the target quantum computing device. This subscription identifier can be generated according to a specific identity identifier of the target quantum computing device or a unique identifier that can identify the target quantum computing device, such as a device identification code of the target quantum computing device.

[0105] Then, the task processing server uses the acquired first subscription identifier to subscribe to the first target message queue in the message proxy server.

[0106] Finally, the task processing server sends the quantum computing subtask to the first target message queue. The first target message queue is a message queue in the message proxy server associated with the target quantum computing device, so the task processing server ensures that the quantum computing subtask can be acquired and processed by the target quantum computing device.

[0107] In this way, through the first subscription identifier and the first target message queue, the task processing server can effectively send the quantum computing subtask to the target quantum computing device and ensure that the quantum computing subtask can be correctly received and processed. At the same time, this message queue-based communication method also provides asynchronous processing capabilities, allowing the task processing server to continue to perform other operations after sending the quantum computing subtask without waiting for the quantum computing subtask to be processed, thereby improving the overall efficiency of the system.

[0108] In some embodiments, the quantum computing power resource device is configured to:

[0109] Get the first subscription identifier;

[0110] Subscribing to a first target message queue according to a first subscription identifier;

[0111] Obtain a quantum computing subtask from a first target message queue;

[0112] The quantum computing subtasks are processed to obtain initial quantum processing results.

[0113] Specifically, first, the quantum computing power resource device also needs to obtain the first subscription identifier. Then, the quantum computing power resource device uses the obtained first subscription identifier to subscribe to the first target message queue in the message proxy server. Then, the quantum computing power resource device obtains the quantum computing subtask from the first target message queue to ensure that the quantum computing subtask obtained from the message proxy server is the task it needs to process. After obtaining the correct quantum computing subtask, the quantum computing power resource device processes the obtained quantum computing subtask to obtain the initial quantum processing result.

[0114] In this way, through the first subscription identifier and the first target message queue, the quantum computing resource device can effectively obtain the quantum computing subtasks to be processed from the message proxy server, and perform corresponding processing to obtain the initial quantum processing results. This communication method based on message queues also provides asynchronous processing capabilities, allowing the quantum computing resource device to continue to monitor the message queue and obtain new tasks while processing quantum computing subtasks.

[0115] In some embodiments, the quantum computing power resource device is configured to:

[0116] Based on the preset security mechanism, the quantum computing subtask is verified to confirm the validity of the source of the quantum computing subtask.

[0117] Specifically, the preset security mechanism refers to a series of pre-set security measures such as sender identity authentication and sender access control. In the implementation of the present application, sender identity authentication, integrity verification of quantum computing subtasks, and validity verification of quantum computing subtasks are used for illustration. The specific security mechanism used in actual applications is determined according to actual needs.

[0118] When sending a quantum computing subtask, the task processing server simultaneously sends the digital certificate or access token of the task processing server, the signature authentication information of the quantum computing subtask, and the timestamp of the quantum computing subtask.

[0119] When the quantum computing resource device obtains the quantum computing subtask, it first verifies the digital certificate or access token in it to ensure that the quantum computing subtask is sent by the task processing server. In addition, by verifying the signature authentication information of the quantum computing subtask, it is confirmed that the quantum computing subtask remains intact during the transmission process and has not been illegally modified. In addition, by checking the timestamp of the quantum computing subtask, it is verified whether the quantum computing subtask is within the validity period and whether it is still valid.

[0120] In this way, based on the preset security mechanism, the quantum computing resource device can ensure that the source of the received quantum computing subtask is valid, the sender is legal, the quantum computing subtask remains intact during the transmission process, and the quantum computing subtask is within the validity period. This helps protect the quantum computing resource device from malicious attacks and illegal access, and ensures the security and reliability of quantum computing.

[0121] In some embodiments, the quantum computing power resource device is configured to:

[0122] Obtaining a second subscription identifier associated with the target quantum computing device;

[0123] Subscribe to a second target message queue according to a second subscription identifier, where the second target message queue is used to indicate a message queue associated with a target quantum computing device among a plurality of second message queues of the message proxy server;

[0124] The initial quantum processing result is optimized based on a preset optimization method to obtain a target quantum processing result, and the target quantum processing result is sent to the second target message queue.

[0125] Specifically, the second target message queue refers to a message queue used to process data information returned from the target quantum computing device to the task processing server.

[0126] The preset optimization methods refer to quantum algorithm optimization, quantum error correction and mitigation, and quantum circuit optimization. Quantum algorithm optimization refers to adjusting the parameters of the quantum algorithm, such as rotation angle, selection of quantum gates, etc., to improve the accuracy and efficiency of the algorithm. Quantum error correction refers to the use of quantum error mitigation techniques, such as surface codes or cat-state coding, to reduce the error rate. Quantum circuit optimization refers to optimizing quantum circuits to reduce the number and depth of quantum gates required to reduce the consumption of computing resources.

[0127] The quantum computing power resource device first needs to obtain the second subscription identifier associated with the target quantum computing device. Then, the quantum computing power resource device can subscribe to the second target message queue according to the second subscription identifier, and the second target message queue is used to indicate the message queue associated with the target quantum computing device in multiple second message queues of the message proxy server. Finally, the quantum computing power resource device can optimize the initial quantum processing result based on a preset optimization method to obtain the target quantum processing result, and send the target quantum processing result to the second target message queue. Optimization refers to further calculation, adjustment or improvement of the initial quantum processing result to improve the accuracy and reliability of the result.

[0128] In this way, through the communication method based on message queues, the ability of asynchronous processing can be provided, so that the quantum computing resource device can continue to listen to the message queue and obtain new tasks while processing tasks.

[0129] In some embodiments, the task processing server is configured to:

[0130] Get the second subscription identifier;

[0131] Subscribing to a second target message queue in the message proxy server according to a second subscription identifier;

[0132] Obtain the target quantum processing result from the second target message queue.

[0133] Specifically, the task processing server first needs to obtain the second subscription identifier. Then, the task processing server can subscribe to the second target message queue in the message proxy server according to the second subscription identifier. Finally, the task processing server can obtain the target quantum processing result from the second target message queue.

[0134] In this way, the communication method based on message queues can provide asynchronous processing capabilities, allowing the task processing server to continue to perform other operations after sending the task without waiting for the task processing to complete. At the same time, this decoupled design also enhances the scalability and robustness of the system, allowing the task processing server to easily interact with multiple quantum computing power resource devices.

[0135] Please refer to Figure 1 ,In some embodiments, the fusion computing system further includes a classic resource access device and a classic computing resource device;

[0136] The task processing server is configured to process the computing task to obtain a classical computing subtask;

[0137] The classic resource access device is configured to call the classic computing resource device to perform computational processing on the classic computing subtasks to obtain the target classic processing results.

[0138] Specifically, the fusion computing system also includes a classic resource access device and a classic computing resource device. The task processing server can process the computing task to obtain a classic computing subtask. The classic resource access device can call the classic computing resource device to perform computing processing on the classic computing subtask to obtain the target classic processing result.

[0139] See also Figure 3 , Figure 3 The figure is a schematic diagram of the classic computing subtask processing flow of the fusion computing system. Among them, the task processing server is the top-level module of the entire scheduling system, responsible for coordinating and managing the workflow of each subsystem and component. It ensures the splitting, allocation and execution of computing tasks, and is also responsible for monitoring the operating status of the entire scheduling system to ensure the stability and reliability of the scheduling system.

[0140] The classic resource access device is located at the second layer. The classic resource access device includes a service API for processing data streams from different sources to ensure the orderly transmission and reception of data. The classic resource access device can realize the unified classic computing resource initialization, classic computing resource application, classic computing resource use and management software layer structure.

[0141] Classic computing resource devices are the third layer, which can realize the following functions: access authorization, credential authentication, credential renewal, user resource management, submitting computing tasks, querying computing tasks, deleting computing tasks and container / file management. Access authorization refers to verifying the identity and permissions of users to ensure that only authorized users can access system resources. Credential authentication refers to verifying the credentials provided by users to protect the security of the system and the integrity of data. Credential renewal refers to managing the validity period of user credentials and reminding users to renew before the credentials expire. User resource management refers to allowing users to manage their own computing resources, such as allocating resources and monitoring resource usage. Submitting computing tasks means that users can submit their own computing tasks through this module, including selecting computing resources and setting task parameters. Querying computing tasks means that users can view the status, progress and results of the tasks they submitted, so as to understand the execution of tasks in a timely manner. Deleting computing tasks means allowing users to cancel or delete the tasks they submitted to release computing resources or avoid unnecessary computing overhead. Container / file management refers to providing file upload, download and management functions, as well as container creation, configuration and management to meet the needs of different computing tasks.

[0142] In this way, the fusion computing system can make full use of classical computing resources and quantum computing resources to achieve efficient processing of computing tasks, and the fusion computing system can flexibly handle various types of computing tasks, improving overall computing power and efficiency.

[0143] See also Figure 4 The embodiment of the present application provides a task processing method, which is based on the above-mentioned fusion computing system and includes:

[0144] 01: The task processing server processes the acquired computing task to obtain the quantum computing subtask, and sends the quantum computing subtask to the message proxy server;

[0145] 02: The message proxy server stores the quantum computing subtasks sent by the task processing server based on the first message queue;

[0146] 03: The quantum computing resource device processes the quantum computing subtasks obtained from the message proxy server, obtains the target quantum processing results, and returns them to the message proxy server;

[0147] 04: Based on the second message queue, the target quantum processing results returned by the quantum computing power resource device are stored;

[0148] 05: The task processing server obtains the target quantum processing result from the message proxy server.

[0149] The embodiment of the present application also provides a computer device, including a memory and a processor. The task processing method of the embodiment of the present application can be implemented by the computer device of the embodiment of the present application. Specifically, a computer program is stored in the memory, and the processor is used to process the acquired computing task to obtain a quantum computing subtask, and send the quantum computing subtask to the message proxy server. And based on the first message queue, the quantum computing subtask sent by the task processing server is stored. And the quantum computing subtask obtained from the message proxy server is processed to obtain the target quantum processing result, and returned to the message proxy server. The processor is also used to store the target quantum processing result returned by the quantum computing power resource device based on the second message queue. And the target quantum processing result is obtained from the message proxy server.

[0150] The task processing method provided in the embodiment of the present application is based on the above-mentioned fusion computing system. Specifically, the task processing server processes the acquired computing task to obtain a quantum computing subtask, and sends the quantum computing subtask to the message proxy server. The message proxy server stores the quantum computing subtask sent by the task processing server based on the first message queue. The quantum computing power resource device processes the quantum computing subtask obtained from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server. Based on the second message queue, the target quantum processing result returned by the quantum computing power resource device is stored. The task processing server obtains the target quantum processing result from the message proxy server.

[0151] Specifically, the task processing server processes the acquired computing task to obtain a quantum computing subtask, and sends the quantum computing subtask to the message proxy server. Then, the message proxy server stores the quantum computing subtask sent by the task processing server based on the first message queue. The quantum computing power resource device then processes the quantum computing subtask acquired from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server. Then, based on the second message queue, the message proxy server stores the target quantum processing result returned by the quantum computing power resource device. Finally, the task processing server obtains the target quantum processing result from the message proxy server.

[0152] In summary, the fusion computing system can make full use of quantum computing resources to achieve efficient processing of computing tasks. The message proxy server can store and forward quantum computing subtasks and quantum processing results to ensure effective communication between the task processing server and the quantum computing resource device. This message queue-based communication method can also provide asynchronous processing capabilities, allowing the task processing server to continue to perform other operations after sending the task without waiting for the task processing to complete. At the same time, through this decoupled design, the scalability and robustness of the system are enhanced, allowing the task processing server to easily interact with multiple quantum computing resource devices.

[0153] See also Figure 5 In some embodiments, the method further comprises:

[0154] 06: The task processing server processes the computing task to obtain the classic computing subtask;

[0155] 07: The classic resource access device calls the classic computing resource device to perform computation on the classic computing subtask and obtain the target classic processing result.

[0156] In some implementations, the task processing server processes the computing task to obtain a classic computing subtask. The classic resource access device calls the classic computing resource device to perform computing processing on the classic computing subtask to obtain a target classic processing result.

[0157] In some embodiments, the processor is further configured to process the computing task to obtain a classical computing subtask, and to call a classical computing resource device to perform computing processing on the classical computing subtask to obtain a target classical processing result.

[0158] Specifically, the task processing server processes the computing task to obtain the classic computing subtask. Then, the classic resource access device calls the classic computing resource device to perform computing processing on the classic computing subtask to obtain the target classic processing result.

[0159] In this way, the fusion computing system can make full use of classic computing resources to achieve efficient processing of computing tasks. Based on classic resource access devices and classic computing resource devices, the system can flexibly handle various types of computing tasks, improving overall computing power and efficiency. At the same time, this message queue-based communication method can also provide asynchronous processing capabilities, allowing the task processing server to continue to perform other operations after sending a task without waiting for the task processing to complete.

[0160] The following example illustrates the above task processing method. The computing task is a precipitation data prediction algorithm based on a hybrid neural network model. Figure 6 , Figure 6 This is a schematic diagram of a hybrid neural network model. The left side is the encoder part, which extracts information through convolution and downsampling; the right side is the decoder part, which recovers information through upsampling. The last convolution layer on the right side is replaced by a quantum neural network. The classical resource access device is the supercomputing data receiving layer. The message proxy server is the quantum computing data receiving layer.

[0161] See also Figure 7 , Figure 7 A quantum computing cloud platform that integrates the fusion computing system provided by the implementation method of the present application. The quantum computing cloud platform includes a front-end client, where users can write and debug quantum computing programs, which is the main interface for users to interact with the quantum computing cloud platform. In addition, users can estimate the required computing resources according to task requirements and select appropriate computing resources for calculation. In addition, users submit the written quantum computing programs and task requirements to the quantum computing cloud platform, and the platform will schedule and execute according to task requirements and resource conditions.

[0162] The quantum computing cloud platform also includes a backend server, which is responsible for receiving the computing task data submitted by the user, and performing preliminary processing and storage. In addition, the backend server submits the processed task data to the fusion computing system, and the scheduling system performs resource allocation and task scheduling. In addition, after the computing task is completed, the backend server returns the result to the front-end client, and the user can view and download the computing results.

[0163] The quantum computing cloud platform also includes a fusion computing system provided in the embodiment of the present application. The fusion computing system can dynamically allocate computing resources according to task requirements and resource conditions to ensure efficient execution of tasks. Moreover, the fusion computing system creates corresponding computing resource instances according to task requirements. In addition, the fusion computing system parses the submitted computing tasks and extracts key information and requirements of the tasks. The fusion computing system splits the computing tasks into multiple subtasks according to the characteristics of the quantum computer and the task requirements, and sends these subtasks to the corresponding receiving layer for subsequent processing.

[0164] The fusion computing system also includes two receiving layers, including a supercomputing data receiving layer and a quantum computing data receiving layer. The supercomputing data receiving layer is used to call classical computing resources to perform computational processing on classical computing subtasks. The quantum computing data receiving layer is used to receive and store the sent quantum computing subtask data information.

[0165] The user submits a precipitation data prediction algorithm based on a hybrid neural network model on the front end of the quantum computing cloud platform and selects computing resources. The selected classical computing resources include 16 central processing unit cores and 64GB of memory, and the quantum computing resources are a superconducting quantum computer with 176 quantum bits. The relevant parameters are only for schematic illustration and may be other parameters in other examples, which are not limited here.

[0166] Next, the backend of the quantum computing cloud platform saves the computing task machine-related data submitted by the user and passes the computing task to the fusion computing system. The fusion computing system creates the computing task instance based on the system default resources and parses and splits the computing task command. The classical computing subtask obtained after the split is then passed to the supercomputing center for processing through the supercomputing data receiving layer, and the quantum computing power resources are called through the quantum computing data receiving layer to process the quantum computing subtask obtained after the split. Finally, the classical computing results and quantum computing results are merged and analyzed to obtain the target processing results, which are then packaged and returned to the quantum computing cloud platform. Users can view and download the target processing results through the quantum computing cloud platform.

[0167] The present application also provides a computer-readable storage medium containing a computer program. When the computer program is executed by one or more processors, the one or more processors execute the method of the present application.

[0168] It is understood that a computer program includes computer program code. The computer program code may be in source code form, object code form, executable file or some intermediate form. Computer readable storage media may include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media.

[0169] In the description of this specification, the descriptions with reference to the terms "specifically", "further", "particularly", "understandably", etc. are intended to mean that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not intended to refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0170] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0171] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A fusion computing system, characterized in that: The fusion computing system includes a task processing server, a message proxy server and a quantum computing power resource device; The task processing server is configured to process the acquired computing task to obtain a quantum computing subtask, and send the quantum computing subtask to the message proxy server; The message proxy server includes a first message queue associated with the quantum computing power resource device, and the message proxy server is configured to store the quantum computing subtask sent by the task processing server based on the first message queue; The quantum computing power resource device is configured to process the quantum computing subtask obtained from the message proxy server, obtain a target quantum processing result, and return it to the message proxy server; The message proxy server includes a second message queue associated with the quantum computing power resource device, and the message proxy server is configured to store the target quantum processing result returned by the quantum computing power resource device based on the second message queue; The task processing server is configured to obtain the target quantum processing result from the message proxy server.

2. The fusion computing system according to claim 1, characterized in that: The task processing server is configured as follows: In response to a resource selection operation, a target quantum computing device is determined from the quantum computing power resource device.

3. The fusion computing system according to claim 2, characterized in that: The task processing server is configured as follows: Obtaining a first subscription identifier associated with the target quantum computing device; subscribing to a first target message queue in the message proxy server according to the first subscription identifier, where the first target message queue is used to indicate a message queue associated with the target quantum computing device among a plurality of first message queues of the message proxy server; Sending the quantum computing subtask to the first target message queue.

4. The fusion computing system according to claim 3, characterized in that: The quantum computing power resource device is configured as follows: Obtaining the first subscription identifier; Subscribing to the first target message queue according to the first subscription identifier; Obtain the quantum computing subtask from the first target message queue; The quantum computing subtask is processed to obtain an initial quantum processing result.

5. The fusion computing system according to claim 4, characterized in that: The quantum computing power resource device is configured as follows: Based on a preset security mechanism, the quantum computing subtask is verified to confirm the validity of the source of the quantum computing subtask.

6. The fusion computing system according to claim 5, characterized in that: The quantum computing power resource device is configured as follows: Obtaining a second subscription identifier associated with the target quantum computing device; Subscribe to a second target message queue according to the second subscription identifier, where the second target message queue is used to indicate a message queue associated with the target quantum computing device among a plurality of second message queues of the message proxy server; The initial quantum processing result is optimized based on a preset optimization method to obtain the target quantum processing result, and the target quantum processing result is sent to the second target message queue.

7. The fusion computing system according to claim 6, characterized in that: The task processing server is configured as follows: Obtaining the second subscription identifier; Subscribing to the second target message queue in the message proxy server according to the second subscription identifier; The target quantum processing result is obtained from the second target message queue.

8. The fusion computing system according to claim 1, characterized in that: The fusion computing system also includes a classic resource access device and a classic computing power resource device; The task processing server is configured to process the computing task to obtain a classic computing subtask; The classic resource access device is configured to call the classic computing resource device to perform computational processing on the classic computing subtask to obtain a target classic processing result.

9. A task processing method, characterized in that: The task processing method is based on the fusion computing system according to claims 1-8, and the method includes: The task processing server processes the acquired computing task to obtain a quantum computing subtask, and sends the quantum computing subtask to the message proxy server; The message proxy server stores the quantum computing subtask sent by the task processing server based on the first message queue; The quantum computing power resource device processes the quantum computing subtask obtained from the message proxy server, obtains the target quantum processing result, and returns it to the message proxy server; The message proxy server stores the target quantum processing result returned by the quantum computing power resource device based on the second message queue; The task processing server obtains the target quantum processing result from the message proxy server.

10. The task processing method according to claim 9, characterized in that: The method further comprises: The task processing server processes the computing task to obtain a classic computing subtask; The classic resource access device calls the classic computing resource device to perform computational processing on the classic computing subtask to obtain a target classic processing result.

Images