Computing power resource monitoring method, excess fusion computing monitoring system and monitoring server

Abstract

The invention discloses a computing power resource monitoring method adaptive to an excess fusion computing monitoring system, the excess fusion computing monitoring system and an excess fusion computing monitoring server. The method comprises: based on a preset interface, according to a target subscription identifier, obtaining target operation data of a target computing device, the target subscription identifier being a subscription identifier subscribed in advance and corresponding to the target operation data; and preprocessing the target operation data to obtain target data. And based on a preset alarm threshold value, monitoring the first operation state of the target computing device according to the target data so as to realize monitoring of computing power resources of the excess fusion computing monitoring system. Therefore, aiming at the characteristics of the excess fusion computing monitoring system, the computing power resources are monitored in real time, and the system is ensured to operate efficiently and stably, so that the effectiveness of algorithm execution in the process of processing a computing task by the system is ensured. And potential performance bottlenecks or faults can be found and processed in time, so that the overall calculation efficiency and the system reliability of the excess fusion calculation monitoring system are optimized.

Description

Technical Field

[0001] The present application relates to the field of quantum computing, and more specifically, to a computing power resource monitoring method adapted for an over-convergent computing monitoring system, an over-convergent computing monitoring system, and an over-convergent computing monitoring server. Background Art

[0002] By combining the strengths of quantum computing and supercomputing, hyperconverged architectures can complement each other in solving complex problems. Related technologies typically use monitoring methods based on classical computer architectures to monitor computing resources within hyperconverged architectures. However, these methods are not suitable for monitoring quantum computing devices. Summary of the Invention

[0003] The present application provides a computing power resource monitoring method adapted to an over-converged computing monitoring system, an over-converged computing monitoring system, and an over-converged computing monitoring server.

[0004] The present application provides a method for monitoring computing resources adapted to an over-converged computing monitoring system, the method comprising:

[0005] Based on a preset interface, obtaining target operating data of a target computing device according to a target subscription identifier, wherein the target subscription identifier is a subscription identifier that is pre-subscribed and corresponds to the target operating data;

[0006] Preprocessing the target operation data to obtain target data;

[0007] Based on a preset alarm threshold, the first operating state of the target computing device is monitored according to the target data to monitor the computing power resources of the hyperconverged computing monitoring system.

[0008] Thus, the embodiment of the present application provides a computing power resource monitoring method adapted to the hyper-converged computing monitoring system. First, based on the preset interface, the hyper-converged computing monitoring system obtains the target operation data of the target computing device according to the target subscription identifier, and the target subscription identifier is a subscription identifier that is pre-subscribed and corresponds to the operation data. Then, the hyper-converged computing monitoring system pre-processes the target operation data to obtain the target data. Finally, based on the preset alarm threshold, the hyper-converged computing monitoring system monitors the first operation state of the target computing device according to the target data to realize the monitoring of the computing power resources of the hyper-converged computing monitoring system. In this way, according to the characteristics of the hyper-converged computing monitoring system, the computing power resources are monitored in real time to ensure that the system can run efficiently and stably, thereby ensuring the effectiveness of the algorithm execution in the process of the system processing computing tasks. And it can timely discover and handle potential performance bottlenecks or faults, thereby optimizing the overall computing efficiency and system reliability of the hyper-converged computing monitoring system.

[0009] In some embodiments, the hyperconverged computing monitoring system includes a plurality of computing devices, and the method further includes:

[0010] A subscription identifier associated with each of the computing devices is set, the subscription identifier including the target subscription identifier.

[0011] Thus, the hyperconverged computing monitoring system includes multiple computing devices, and a subscription identifier is associated with each computing device. The subscription identifier includes a target subscription identifier. By setting a subscription identifier associated with each computing device, the hyperconverged computing monitoring system achieves asynchronous and decoupled data transmission, enhancing the scalability and robustness of the system.

[0012] In some embodiments, obtaining target operating data of a target computing device based on a preset interface and a target subscription identifier includes:

[0013] Based on the preset interface, obtaining and storing the operating data of all computing devices in the hyperconvergent computing monitoring system, the operating data including the performance indicators and operating logs of each computing device;

[0014] The target operating data is obtained from the operating data according to the target subscription identifier.

[0015] First, based on a pre-set interface, the hyperconverged computing monitoring system acquires and stores the operational data of all computing devices in the hyperconverged computing monitoring system. This operational data includes the performance indicators and operational logs of each computing device. Next, the hyperconverged computing monitoring system retrieves target operational data from the operational data based on the target subscription identifier. This acquisition of target operational data based on the target subscription identifier enables subsequent evaluation of the target computing device's status and performance.

[0016] In certain embodiments, preprocessing the target operating data to obtain target data includes:

[0017] Analyzing the target operation data to obtain target analytical data;

[0018] Converting the target parsed data to obtain target format data;

[0019] The target format data is filtered to obtain target data.

[0020] Thus, first, the hyperconverged computing monitoring system parses and processes the target operating data to obtain target parsed data, that is, extracts useful information from the target operating data. Next, the hyperconverged computing monitoring system converts and processes the target parsed data to obtain target format data, that is, converts the parsed data into target format data with a unified format to facilitate subsequent analysis and processing. Finally, the hyperconverged computing monitoring system filters and processes the target format data to obtain target data, that is, removes irrelevant or unnecessary information from the target format data, retaining only the target data that is useful for the current analysis or monitoring task. In this way, the hyperconverged computing monitoring system can ensure that only the most relevant data is processed and analyzed, improving the efficiency and accuracy of data processing.

[0021] In certain embodiments, the method further comprises:

[0022] The historical operation data of the target computing device is updated according to the target data to predict characteristic data of the target computing device according to the historical operation data, wherein the characteristic data includes performance trends and potential failure points of the target computing device.

[0023] In this way, the hyperconverged computing monitoring system updates the historical operating data of the target computing device based on the target data, predicting the target computing device's characteristic data based on this historical operating data. This characteristic data includes the target computing device's performance trends and potential failure points. By integrating new performance indicators and operating logs into existing historical operating data, the timeliness and integrity of the historical operating data are maintained. Furthermore, by utilizing this historical operating data to predict the characteristic data of the target computing device, the hyperconverged computing monitoring system maintains its stability and reliability, ensuring efficient system operation.

[0024] In some embodiments, monitoring the first operating state of the target computing device according to the target data based on a preset alarm threshold includes:

[0025] When the target data meets the preset alarm threshold, confirming that the first operating state is normal;

[0026] When the target data does not meet the preset alarm threshold, it is confirmed that the first operating state is abnormal.

[0027] In this way, first, if the target data meets the preset alarm threshold, the first operating state is confirmed to be normal. Then, if the target data does not meet the preset alarm threshold, the first operating state is confirmed to be abnormal. In this way, based on the monitoring and evaluation of the preset alarm thresholds, the hyperconverged computing monitoring system can promptly detect and respond to abnormal states of the target computing device, and take appropriate measures to ensure system stability and performance.

[0028] In certain embodiments, the method further comprises:

[0029] If the first operating state is normal, visualization data is generated according to the target data, and the target data is displayed according to the visualization data.

[0030] In this way, if the first operating state is normal, visualization data is generated based on the target data, and the target data is displayed based on the visualization data. In this way, by displaying the target data based on the visualization data, the hyperconverged computing monitoring system can help users quickly understand and analyze the operating state of the equipment, promptly identify any possible trends or patterns, and take appropriate measures to maintain or improve system performance.

[0031] In certain embodiments, the method further comprises:

[0032] If the first operating state is abnormal, calibrating the target computing device within a preset calibration number to obtain resource calibration data;

[0033] Based on the preset alarm threshold, the second operating state of the target computing device is monitored according to the resource calibration data to determine the operating state of the target computing device after the calibration process is performed.

[0034] Thus, first, if the first operating state is abnormal, the target computing device is calibrated within a preset calibration number of times to obtain resource calibration data. Next, based on a preset alarm threshold, the second operating state of the target computing device is monitored according to the resource calibration data to determine the operating state of the target computing device after the calibration process. In this way, through timely calibration and monitoring, the hyperconverged computing monitoring system can effectively handle anomalies, reduce downtime, and improve overall performance, thereby ensuring the stability and reliability of computing devices in the hyperconverged computing monitoring system.

[0035] In some embodiments, monitoring the second operating status data of the target computing device according to the resource calibration data based on the preset alarm threshold and the preset calibration number includes:

[0036] If the target computing device operates abnormally after the calibration process within a preset number of calibrations, a calibration work order is generated, and the calibration work order is used to indicate the abnormal operating status of the target computing device.

[0037] If the target computing device exhibits abnormal operation after calibration within a preset number of calibration cycles, the hyperconvergent computing monitoring system generates a calibration work order, which indicates the abnormal operating status of the target computing device. This automated repair and step-by-step guidance allows users to quickly correct errors, reducing debugging time, effectively improving the accuracy of their quantum circuit and instruction code designs, and enhancing the user development experience.

[0038] The present application provides a hyperconvergent computing monitoring system, which includes a data collection module, a data analysis module, and a data processing module.

[0039] The data collection module is configured to obtain target operating data of a target computing device according to a target subscription identifier based on a preset interface, wherein the target subscription identifier is a subscription identifier that is pre-subscribed and corresponds to the operating data;

[0040] The data analysis module is configured to pre-process the target operation data to obtain target data;

[0041] The data processing module is configured to monitor the first operating state of the target computing device according to the target data based on a preset alarm threshold, so as to monitor the computing power resources of the hyper-converged computing monitoring system.

[0042] Thus, the embodiment of the present application provides a hyperconvergent computing monitoring system, which includes a data collection module, a data analysis module and a data processing module. The data collection module can obtain the target operation data of the target computing device according to the target subscription identifier based on the preset interface, and the target subscription identifier is a pre-subscribed subscription identifier corresponding to the operation data. The data analysis module can pre-process the target operation data to obtain the target data. The data processing module can monitor the first operating state of the target computing device according to the target data based on the preset alarm threshold to realize the monitoring of the computing power resources of the hyperconvergent computing monitoring system. In this way, based on the low-code platform, it can ensure that users can design and modify quantum circuits through simple graphical operations or quantum instructions, and the low-code platform is responsible for converting these operations into efficient and executable quantum instructions. This method not only lowers the threshold for quantum programming, but also improves development efficiency and promotes the application and development of quantum computing technology.

[0043] An embodiment of the present application provides an over-convergent computing monitoring server. The adapted over-convergent computing monitoring server includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the above method is implemented.

[0044] 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 obvious from the description below, or will be learned through practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] 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:

[0046] Figure 1 This is one of the flow charts of the computing power resource monitoring method of certain embodiments of the present application;

[0047] Figure 2 is a schematic diagram of the architecture of a hyperconvergent computing monitoring system according to certain embodiments of the present application;

[0048] Figure 3 This is the second flow chart of the computing power resource monitoring method in certain embodiments of the present application;

[0049] Figure 4 This is the third flow chart of the computing power resource monitoring method in certain embodiments of the present application;

[0050] Figure 5 This is the fourth flow chart of the computing power resource monitoring method of certain embodiments of the present application;

[0051] Figure 6 This is the fifth flow chart of the computing power resource monitoring method of certain embodiments of the present application;

[0052] Figure 7 This is the sixth flow chart of the computing power resource monitoring method in certain embodiments of the present application;

[0053] Figure 8 This is the seventh flow chart of the computing power resource monitoring method in certain embodiments of the present application;

[0054] Figure 9 This is the eighth flow chart of the computing power resource monitoring method of certain embodiments of the present application;

[0055] Figure 10 This is the ninth flow chart of the computing power resource monitoring method according to certain embodiments of the present application;

[0056] Figure 11 is a schematic diagram of the structure of an over-convergent computing monitoring system according to certain embodiments of the present application;

[0057] Figure 12 This is a structural diagram of an example of a super-convergent computing monitoring system for certain embodiments of the present application. DETAILED DESCRIPTION

[0058] The embodiments of the present application are described in detail below. 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 throughout. 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 should not be understood as limiting the embodiments of the present application.

[0059] By combining the strengths of quantum computing and supercomputing, hyperconverged architectures can complement each other in solving complex problems. Quantum computing's powerful parallel processing capabilities can accelerate the solution of certain types of problems, particularly those that are extremely complex or difficult to solve using classical computing. Meanwhile, the efficient numerical computation capabilities of supercomputers can handle a large number of traditional computing tasks, ensuring system stability and reliability.

[0060] However, related technologies typically use monitoring methods based on classical computer architectures to monitor computing resources in hyperconverged architectures. This monitoring approach primarily focuses on performance metrics of traditional computing resources, such as CPU utilization, memory usage, storage performance, and network latency. These metrics are not applicable to the operation of quantum computers, as quantum computers are centered around quantum properties such as qubits, quantum gate operations, and quantum decoherence, parameters that are not incorporated into traditional monitoring schemes.

[0061] Furthermore, the operation of quantum computers relies on a highly sophisticated hardware environment, particularly physical conditions such as low temperature and high vacuum, as well as error correction mechanisms. Traditional monitoring tools are not designed to track the connection between this specialized physical environment and the core operations of quantum computing, making it difficult to monitor key metrics such as qubit decoherence time and gate operation accuracy. Therefore, an operational monitoring solution tailored to this hybrid architecture is urgently needed to effectively improve system stability and operational efficiency.

[0062] Based on the above questions, please refer to Figure 1 The present application provides a method for monitoring computing resources adapted to the hyperconvergent computing monitoring system 100, the method comprising:

[0063] 01: Based on the preset interface, obtain the target operation data of the target computing device according to the target subscription identifier;

[0064] 02: Preprocess the target operation data to obtain target data;

[0065] 03: Based on the preset alarm threshold, the first operating state of the target computing device is monitored according to the target data to monitor the computing power resources of the hyper-converged computing monitoring system.

[0066] The embodiment of the present application also provides an over-converged computing monitoring server, including a memory and a processor. The computing power resource monitoring method adapted to the over-converged computing monitoring system of the embodiment of the present application can be implemented by the over-converged computing monitoring server of the embodiment of the present application. Specifically, a computer program is stored in the memory, and the processor is used to obtain the target operation data of the target computing device according to the target subscription identifier based on the preset interface. And pre-process the target operation data to obtain the target data. And based on the preset alarm threshold, the first operation status data of the target computing device is monitored according to the target data to realize the monitoring of the computing power resources of the over-converged computing monitoring system.

[0067] The embodiment of the present application also provides a hyper-converged computing monitoring system 100. The computing power resource monitoring method adapted to the hyper-converged computing monitoring system 100 of the embodiment of the present application can be implemented by the hyper-converged computing monitoring system 100 of the embodiment of the present application. Specifically, the hyper-converged computing monitoring system 100 includes a data collection module 10 and a data analysis module 20. The data collection module 10 is used to obtain the target operation data of the target computing device according to the target subscription identifier based on a preset interface. The data analysis module 20 is used to pre-process the target operation data to obtain the target data. And based on the preset alarm threshold, the first operation status data of the target computing device is monitored according to the target data to realize the monitoring of the computing power resources of the hyper-converged computing monitoring system 100.

[0068] Specifically, super-convergent computing refers to a new computing architecture that combines the parallel processing capabilities of quantum computing with the efficient numerical computing capabilities of supercomputers.

[0069] The hyperconvergent computing monitoring system 100 is a service system based on a hyperconvergent computing architecture. The hyperconvergent computing monitoring system 100 can adapt to different computing needs and improve computing efficiency and performance by integrating different types of computing resources and technologies, such as traditional central processing units (CPUs), graphics processing units (GPUs), tensor processing units (TPUs), and quantum processing units such as qubits in quantum computers.

[0070] High-Performance Computing (HPC) refers to computer systems designed to perform complex computational tasks, boasting extremely high computing power, storage capacity, and data processing speed. HPCs are widely used in scientific research, engineering simulation, data analysis, weather forecasting, bioinformatics, and other fields.

[0071] A graphics processing unit (GPU) is a microprocessor specifically designed for handling graphics and image-related computations. It includes numerous computing cores and can handle multiple tasks simultaneously. GPUs also have high memory bandwidth, enabling them to quickly read and write large amounts of data, giving them a natural advantage in processing image and video data.

[0072] A tensor processing unit (TPU) is a specialized integrated circuit (ASIC) designed specifically for machine learning and deep learning tasks. It aims to optimize and accelerate neural network computations, particularly for training and running deep learning models. A TPU can handle computations across multiple neural network layers in parallel, accelerating model training and inference. Furthermore, it delivers higher performance while consuming less power when executing deep learning tasks.

[0073] Pre-defined interfaces are pre-defined interfaces used for interaction and communication between the hyperconverged computing monitoring system 100 and computing devices. These interfaces specify the format, protocol, and standards for data exchange, ensuring that key performance indicators (KPIs) of computing devices, i.e., their operational data, are accurately collected and sent to the hyperconverged computing monitoring system 100.

[0074] Subscription identifiers refer to tags or key values ​​used to identify different computing devices. By using these subscription identifiers, the hyperconverged computing monitoring system 100 can ensure that subscribers or consumers only receive data they are interested in, thereby improving the efficiency and accuracy of data processing.

[0075] The target subscription identifier refers to a tag or key value used to identify the target computing device.

[0076] The target computing device refers to a computing device of interest specified by a user or other input device in the computing resources of the hyperconverged computing monitoring system 100 .

[0077] Target operation data refers to the operation data of the target computing device when processing tasks. This operation data may include performance indicators and log records of the computing device.

[0078] The target data refers to the data information that can be directly analyzed after the hyper-converged computing monitoring system 100 processes the acquired target operation data.

[0079] Preset alarm thresholds refer to a set of specific thresholds or conditions set for computing resources in the hyperconverged computing monitoring system 100. When certain performance indicators or status of computing resources exceed these thresholds, the system will trigger an alarm or alert. These thresholds are pre-set based on expectations of the normal operation of computing resources and the identification of potential problems.

[0080] The computing power resources of the hyperconverged computing monitoring system 100 refer to the hardware and software resources that can perform computing tasks, including traditional central processing units (CPUs), graphics processing units (GPUs), tensor processing units (TPUs), and quantum processing units such as quantum bits in quantum computers.

[0081] See also Figure 2 , Figure 2 The following is a schematic diagram of the architecture of the hyperconvergent computing monitoring system 100. The hyperconvergent computing monitoring system 100 includes a data source module 50, a data collection module 10, a data analysis module 20, a data processing module 30, and a data display module 40. The data source module 50 lists the data sources that the data monitoring and analysis system relies on. The data source module 50 mainly includes three core devices: a quantum computer, a quantum simulator, and a high-performance supercomputer. During operation, a quantum computer generates a large amount of performance indicators and system logs, such as single-bit gate fidelity, single-bit coherence time, two-bit gate fidelity, single-bit read fidelity, read fidelity, etc., as well as the computational accuracy of quantum algorithms, qubit states, execution time, and other hardware-related parameters. When the quantum simulator is in the operation state of simulating a quantum computer, it also generates performance indicator data such as the computational accuracy of quantum algorithms, qubit states, execution time, and system logs. High-performance supercomputers can provide powerful computing resources for quantum computing. While supporting quantum algorithm simulation and complex computing task processing, they will also generate performance indicator data information and system logs such as the quantum algorithm's computing accuracy, quantum bit status, and execution time.

[0082] The data collection module 10 acquires and stores data such as system logs and performance metrics from the data source module 50. It transmits this data to subsequent modules via APIs or log file parsing. APIs are interfaces that enable data transmission, while log file parsing refers to the process of extracting, interpreting, and converting data from system logs. System logs record the operation history of computing devices, such as quantum computing devices, including operation records, system status, and error logs. Performance metrics include qubit fidelity, quantum algorithm computational accuracy, qubit state, execution time, and other hardware-related parameters.

[0083] The data analysis module 20 can conduct in-depth analysis of the collected data. By performing real-time analysis on the data of the data collection module 10 and judging the current performance status of the quantum computer based on the preset alarm threshold, the performance status and health status of the quantum computer can be obtained.

[0084] The data processing module 30 processes the analysis results and may trigger corresponding actions. It primarily consists of two components: automated scripts and alarm notifications. When the system detects performance indicators deviating from standards, i.e., exceeding preset alarm thresholds, the automated scripts automatically trigger a series of scripts for calibration and repair, minimizing manual intervention. When performance indicators exceed warning thresholds, the alarm notifications send calibration work orders via email, text message, or other means to notify relevant technicians, ensuring timely resolution of the issue.

[0085] The data display module 40 visualizes processed data for operations and maintenance personnel, helping them better understand the system's current status. The engineering tracking module displays the system's operation logs and overall operational status. The quantum dashboard dynamically displays key performance indicators related to the quantum computer, facilitating real-time monitoring.

[0086] It should be noted that in the embodiments of the present application, performance indicator data and system logs and other data information generated by classical computing devices are monitored by a monitoring method based on classical computer architecture.

[0087] First, the data collection module 10 obtains the target operating data of the target computing device according to the target subscription identifier through the preset interface, and the hyper-converged computing monitoring system 100. Then, the data analysis module 20 pre-processes the acquired target operating data to obtain the target data. The purpose of pre-processing is to clean, convert and organize the data so that it is suitable for subsequent analysis and monitoring. Finally, the data analysis module 20 monitors the first operating state of the target computing device according to the target data based on the preset alarm threshold. If the target data exceeds the preset alarm threshold, the system will trigger an alarm, thereby realizing the monitoring of the computing power resources of the hyper-converged computing monitoring system 100.

[0088] In summary, the embodiments of the present application provide a computing power resource monitoring method adapted for an over-converged computing monitoring system 100, an over-converged computing monitoring system 100, and an over-converged computing monitoring server. First, based on a preset interface, the over-converged computing monitoring system 100 obtains the target operating data of the target computing device according to the target subscription identifier, where the target subscription identifier is a pre-subscribed subscription identifier corresponding to the operating data. Next, the over-converged computing monitoring system 100 pre-processes the target operating data to obtain the target data. Finally, based on a preset alarm threshold, the over-converged computing monitoring system 100 monitors the first operating state of the target computing device according to the target data to achieve monitoring of the computing power resources of the over-converged computing monitoring system 100. In this way, based on the characteristics of the over-converged computing monitoring system 100, the computing power resources are monitored in real time to ensure that the system can operate efficiently and stably, thereby ensuring the effectiveness of algorithm execution during the system processing computing tasks. It can also promptly discover and handle potential performance bottlenecks or faults, thereby optimizing the overall computing efficiency and system reliability of the over-converged computing monitoring system 100.

[0089] See also Figure 3 In some embodiments, the hyperconverged computing monitoring system includes a plurality of computing devices, and the method further includes:

[0090] 04: Set the subscription ID associated with each computing device.

[0091] In some embodiments, the data collection module 10 is further configured to set a subscription identifier associated with each computing device.

[0092] In some embodiments, the processor is further configured to set a subscription identifier associated with each computing device.

[0093] Specifically, the data collection module 10 assigns subscription identifiers to the multiple computing devices in the hyperconverged computing monitoring system 100. These subscription identifiers can be derived from the device's physical address, logical identifier, or any other uniquely identifiable code. By associating the subscription identifiers with the computing devices, when the data collection module 10 collects data, it can identify and classify the data based on the subscription identifiers, ensuring that the data corresponds to the correct devices and services.

[0094] In this way, by setting a subscription identifier associated with each computing device, the hyper-converged computing monitoring system 100 achieves asynchrony and decoupling of data transmission, thereby enhancing the scalability and robustness of the system.

[0095] See also Figure 4 In some embodiments, step 01 (obtaining target operating data of a target computing device according to a target subscription identifier based on a preset interface) includes:

[0096] 011: Based on the preset interface, obtain and store the operating data of all computing devices in the hyper-converged computing monitoring system;

[0097] 012: Get target running data from running data according to the target subscription ID.

[0098] In some embodiments, the data collection module 10 is further configured to obtain and store the operating data of all computing devices in the hyperconvergent computing monitoring system based on a preset interface, and to obtain target operating data from the operating data according to a target subscription identifier.

[0099] In some embodiments, the processor is further configured to obtain and store the operating data of all computing devices in the hyperconvergent computing monitoring system based on a preset interface, and obtain target operating data from the operating data according to the target subscription identifier.

[0100] Specifically, the hyperconverged computing monitoring system 100 first acquires operational data from all computing devices in the system through a pre-set interface. This data includes performance metrics and operational logs for each computing device, which are crucial for assessing device status and performance. Furthermore, the acquired data is stored in the system's data storage facilities for subsequent analysis and processing.

[0101] Then, after storing the operating data of all computing devices, the hyperconverged computing monitoring system 100 extracts target operating data from the operating data according to the target subscription identifier. In this way, the hyperconverged computing monitoring system 100 can focus on analyzing specific devices, improving the pertinence and efficiency of monitoring.

[0102] In this way, target operating data is obtained from the operating data according to the target subscription identifier, and the state and performance of the target computing device can be evaluated in a subsequent process based on the target operating data.

[0103] See also Figure 5 In some embodiments, step 02 (preprocessing the target operating data to obtain target data) includes:

[0104] 021: Analyze and process the target operation data to obtain target analysis data;

[0105] 022: Convert the target parsed data to obtain the target format data;

[0106] 023: Filter the target format data to obtain the target data.

[0107] In certain embodiments, the data analysis module 20 includes a data parsing submodule, a data format conversion submodule, and a data filtering submodule. The data parsing submodule is configured to parse the target operating data to obtain target parsed data. The data format conversion submodule is configured to convert the target parsed data to obtain target format data. The data filtering submodule is configured to filter the target format data to obtain target data.

[0108] In certain embodiments, the processor is further configured to parse the target operating data to obtain target parsed data, convert the target parsed data to obtain target format data, and filter the target format data to obtain target data.

[0109] Specifically, the hyperconvergent computing monitoring system 100 first parses the target operational data to obtain target parsed data. This allows for the extraction of useful information from the raw data, namely the acquired performance data and operational logs. For example, the system parses performance indicator data to identify key performance parameters such as qubit fidelity, quantum algorithm computational accuracy, qubit state, execution time, and other hardware-related parameters.

[0110] Next, the system converts the parsed target data to obtain target format data, that is, converting the parsed data into a unified format to facilitate subsequent analysis and processing. For example, the system may convert data in different formats generated by computing devices into the same format for comparison and analysis.

[0111] Finally, the hyperconverged computing monitoring system 100 filters the converted target format data to obtain target data. This involves removing irrelevant or unnecessary information from the data, retaining only data useful for the current analysis or monitoring task. For example, the system may filter out performance data generated by classical computers based on specific rules or thresholds, retaining only performance data generated by quantum computers.

[0112] In this way, the hyperconverged computing monitoring system 100 can ensure that only the most relevant data is processed and analyzed, thereby improving the efficiency and accuracy of data processing.

[0113] See also Figure 6 In certain embodiments, the method further comprises:

[0114] 05: Update the historical operation data of the target computing device according to the target data to predict the characteristic data of the target computing device according to the historical operation data.

[0115] In some embodiments, the data analysis module 20 further includes a data update submodule, which is used to update the historical operation data of the target computing device according to the target data to predict the characteristic data of the target computing device according to the historical operation data.

[0116] In some embodiments, the processor is further configured to update historical operating data of the target computing device according to the target data, so as to predict characteristic data of the target computing device according to the historical operating data.

[0117] Specifically, performance trends refer to the performance of the target computing device over a period of time in the future, which can help users plan and optimize resource allocation.

[0118] Potential failure points are potential failure points on target computing devices. By informing technicians of these potential failure points, maintenance or replacement can be performed in advance, avoiding unexpected downtime and data loss.

[0119] Historical operation data refers to the operation data of the target computing device collected over a period of time in the past. The historical operation data can be used to predict the characteristic data of the target computing device.

[0120] The hyperconverged computing monitoring system 100 uses the newly acquired target data to update the historical operation data of the target computing device, that is, to integrate the new performance indicators and operation logs into the existing historical data set to maintain the timeliness and integrity of the operation data.

[0121] Next, after updating historical operating data, the system uses this data to predict the characteristic data of the target computing device. Characteristic data primarily refers to the device's performance trends and potential failure points. The prediction process uses methods such as statistical analysis, machine learning models, or time series analysis to identify changing trends in device performance and potential failure risks.

[0122] In this way, by integrating new performance indicators and operation logs into existing historical operation data, the timeliness and integrity of the historical operation data are maintained. Furthermore, by using this historical operation data to predict the characteristic data of the target computing device, the stability and reliability of the hyperconverged computing monitoring system 100 can be maintained, ensuring the efficient operation of the system.

[0123] See also Figure 7 In some embodiments, step 03 (monitoring the first operating state of the target computing device according to the target data based on a preset alarm threshold) includes:

[0124] 031: When the target data meets the preset alarm threshold, confirm that the first operating state is normal;

[0125] 032: When the target data does not meet the preset alarm threshold, it is confirmed that the first operating state is abnormal.

[0126] In some embodiments, the hyperconvergent computing monitoring system 100 further includes a data processing module 30, configured to confirm that the first operating state is normal when the target data meets a preset alarm threshold. The data processing module 30 is configured to confirm that the first operating state is abnormal when the target data does not meet the preset alarm threshold.

[0127] In some embodiments, the processor is further configured to confirm that the first operating state is normal when the target data meets a preset alarm threshold, and to confirm that the first operating state is abnormal when the target data does not meet the preset alarm threshold.

[0128] Specifically, the preset alarm thresholds include a single-bit gate fidelity threshold (99.5%), a single-bit coherence time (20 μs), a dual-bit gate fidelity threshold (96%), and a single-bit read fidelity threshold (90%). It should be noted that the above preset alarm thresholds can be modified by the user as needed.

[0129] The hyperconverged computing monitoring system 100 continuously monitors the first operating state of the target computing device. If the target data of the target computing device meets the preset alarm threshold, that is, the data is within the normal range, the system will confirm that the first operating state is normal, that is, the target computing device is operating normally. This indicates that the performance and functions of the target computing device meet expectations and no abnormalities requiring attention have occurred.

[0130] If the operating status data does not meet the preset alarm threshold, meaning the data falls outside the normal range or exhibits an abnormal pattern, the system will confirm that the device is operating abnormally. For example, if the fidelity of the target computing device drops below 99% when executing a single-bit gate operation due to environmental interference with the qubit or a defect in the device itself, falling below the set 99.5% threshold, this indicates that the target computing device has experienced a certain abnormality and requires attention.

[0131] Due to noise during the reading process or device instability, the fidelity of the device drops below 85% when performing a single-bit read operation, which is lower than the set 90% threshold. This indicates that the target computing device has some abnormalities and needs to be processed.

[0132] Due to interference in the interaction between the two quantum bits or defects in the device itself, the fidelity of the device dropped below 94% when performing a two-bit gate operation, which is lower than the set threshold of 96%. This indicates that the target computing device has certain abnormalities and needs to be processed.

[0133] Due to the interaction between quantum bits and the environment, when the device performs single-bit operations, the single-bit coherence time drops below 15us, which is lower than the set 20us threshold. This indicates that the target computing device has certain abnormalities and needs to be processed.

[0134] In this way, based on the monitoring and evaluation of preset alarm thresholds, the hyper-converged computing monitoring system 100 can promptly detect and respond to abnormal conditions of target computing devices, thereby taking corresponding measures to ensure the stability and performance of the system.

[0135] See also Figure 8 In certain embodiments, the method further comprises:

[0136] 06: If the first operating state is normal, generate visualization data according to the target data, and display the target data according to the visualization data.

[0137] In some embodiments, the hyperconverged computing monitoring system 100 further includes a data display module 40 , which is configured to generate visualization data based on the target data and display the target data based on the visualization data if the first operating state is normal.

[0138] In some embodiments, the processor is further configured to generate visualization data based on the target data if the first operating state is normal, and display the target data based on the visualization data.

[0139] Specifically, when the target computing device is operating normally, the hyperconverged computing monitoring system 100 uses the target data to generate visualization data. This involves converting performance indicators and operation logs into charts, dashboards, or other graphical formats to more intuitively display the device's operating status and trends. After generating the visualization data, the system displays the target data based on this data, specifically through a graphical user interface (GUI) or other display tool, allowing users to view the target computing device's performance indicators, such as real-time monitoring charts and historical trend analysis.

[0140] In this way, by displaying target data based on visual data, the hyper-converged computing monitoring system 100 can help users quickly understand and analyze the operating status of the equipment, promptly discover any possible trends or patterns, and take appropriate measures to maintain or improve system performance.

[0141] See also Figure 9 In certain embodiments, the method further comprises:

[0142] 07: If the first operating state is abnormal, calibrate the target computing device within a preset calibration number to obtain resource calibration data;

[0143] 08: Based on a preset alarm threshold, the second operating state of the target computing device is monitored according to the resource calibration data to determine the operating state of the target computing device after the calibration process.

[0144] In certain embodiments, the quantum processing module is further configured to, if the first operating state is abnormal, perform calibration processing on the target computing device within a preset calibration number to obtain resource calibration data. Furthermore, based on a preset alarm threshold, monitor the second operating state of the target computing device according to the resource calibration data to determine the operating state of the target computing device after the calibration processing.

[0145] In some embodiments, the processor is further configured to, if the first operating state is abnormal, perform calibration on the target computing device within a preset calibration number of times to obtain resource calibration data, and, based on a preset alarm threshold, monitor the second operating state of the target computing device according to the resource calibration data to determine the operating state of the target computing device after the calibration.

[0146] Specifically, the preset calibration times refers to the maximum calibration times set by a user or other input device.

[0147] The second operating state refers to the operating state of the target computing device after the target computing device performs calibration processing.

[0148] When a target computing device experiences an abnormal operation, the hyperconverged computing monitoring system 100 will calibrate the device within a preset number of calibration cycles. Calibration can adjust the configuration or parameters of the target computing device to restore normal operation. Calibration may include hardware adjustments, software updates, system reboots, or configuration changes.

[0149] After the calibration process, the system obtains resource calibration data, which reflects the effect of the calibration operation and the current resource status of the device. The system then uses the resource calibration data to monitor the second operating state of the target computing device based on a preset alarm threshold.

[0150] By monitoring resource calibration data, the system can determine the operating status of the target computing device after the calibration process. If the device's performance indicators return to normal, the system will confirm that the calibration was successful. If the performance problem persists, the system may take further measures, such as repeating the calibration or issuing an alert.

[0151] In this way, through timely calibration and monitoring, the hyperconverged computing monitoring system 100 can effectively handle exceptions, reduce downtime, and improve overall performance, thereby ensuring the stability and reliability of computing devices in the hyperconverged computing monitoring system 100.

[0152] See also Figure 10In some embodiments, step 08 (monitoring the second operating status data of the target computing device according to the resource calibration data based on the preset alarm threshold and the preset calibration number) includes:

[0153] 081: If the target computing device operates abnormally after calibration within the preset number of calibrations, a calibration work order is generated. The calibration work order is used to indicate the abnormal operating status of the target computing device.

[0154] In some embodiments, the quantum processing module is further configured to generate a calibration work order if the target computing device operates abnormally after calibration within a preset number of calibration times. The calibration work order is configured to indicate the abnormal operating status of the target computing device.

[0155] In some embodiments, the processor is configured to generate a calibration work order in response to the target computing device operating abnormally after calibration within a preset number of calibrations, where the calibration work order is configured to indicate the abnormal operating state of the target computing device.

[0156] Specifically, after the calibration process, the hyperconvergent computing monitoring system 100 continuously monitors the operating status of the target computing device. If the target computing device still operates abnormally after the calibration process, that is, the performance indicators do not meet the preset alarm thresholds, the system will record these abnormal conditions.

[0157] If the target computing device behaves abnormally after calibration within a preset number of calibrations, the hyperconvergent computing monitoring system 100 will generate a calibration work order. The calibration work order can be a document or record that indicates the abnormal operating status of the target computing device and provides detailed information about the calibration attempts and results. Relevant technicians use the calibration work order to further investigate and resolve the abnormal device status. Solutions include recalibrating the current state of the physical machine through a series of physical experiments and continuously changing the environment and state of the quantum computer to establish a suitable computer state.

[0158] In this way, through automatic repair and step-by-step guidance, users can quickly correct errors and reduce debugging time, which can effectively improve the correctness of user-designed quantum circuits and quantum instruction codes and enhance the user development experience.

[0159] See also Figure 11 The embodiment of the present application provides a hyperconvergent computing monitoring system 100 , which includes a data collection module 10 , a data analysis module 20 and a data processing module 30 .

[0160] Specifically, the data collection module 10 is configured to obtain target operating data of a target computing device according to a target subscription identifier based on a preset interface, where the target subscription identifier is a pre-subscribed subscription identifier corresponding to the operating data.

[0161] The data analysis module 20 is configured to pre-process the target operation data to obtain target data.

[0162] The data processing module 30 is configured to monitor the first operating state of the target computing device according to the target data based on a preset alarm threshold, so as to monitor the computing power resources of the hyper-converged computing monitoring system 100 .

[0163] In summary, the embodiment of the present application provides an ultra-convergent computing monitoring system 100, which includes a data collection module 10, a data analysis module 20 and a data processing module 30. The data collection module 10 can obtain the target operation data of the target computing device based on the target subscription identifier based on a preset interface, and the target subscription identifier is a pre-subscribed subscription identifier corresponding to the operation data. The data analysis module 20 can pre-process the target operation data to obtain the target data. The data processing module 30 can monitor the first operating state of the target computing device according to the target data based on the preset alarm threshold to realize the monitoring of the computing power resources of the ultra-convergent computing monitoring system 100. In this way, based on the low-code platform, it can ensure that users can design and modify quantum circuits through simple graphical operations or quantum instructions, and the low-code platform is responsible for converting these operations into efficient and executable quantum instructions. This method not only lowers the threshold for quantum programming, but also improves development efficiency and promotes the application and development of quantum computing technology.

[0164] An embodiment of the present application provides an over-convergent computing monitoring server. The adapted over-convergent computing monitoring server includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the above method is implemented.

[0165] The following is a complete description of the computing resource monitoring method adapted for the hyperconvergent computing monitoring system 100 using an example. Figure 12 , Figure 12 The figure is a schematic diagram of the architecture of the specific implementation of the hyperconverged computing monitoring system 100. Kafka is an open-source distributed event stream processing platform that is widely used in high-performance data pipelines, streaming analysis, data integration, and mission-critical applications, enabling high-throughput, low-latency data transmission.

[0166] Logstash is an open source data collection engine with real-time pipeline capabilities. It can collect data from multiple sources, filter and format the data, and then transmit it to the target system.

[0167] ElasticSearch is a distributed, RESTful search and data analytics engine, as well as a scalable data store and vector database. It centrally stores data, enabling fast searches, sophisticated relevance tuning, and powerful analytics, making it suitable for processing large amounts of structured and unstructured data.

[0168] OSS (Object Storage Service) refers to a cloud-based storage solution that stores, manages, and accesses massive amounts of unstructured data in the form of objects, such as images, videos, backup files, and logs.

[0169] Apache Spark, or simply Spark for short, is an open-source distributed computing framework designed for processing large datasets. It offers significantly faster processing speeds than traditional big data processing frameworks and is widely used in big data analysis, data mining, and real-time data processing.

[0170] Kibana is a data visualization tool that provides powerful charting and dashboarding capabilities, primarily designed for use with Elasticsearch. It allows users to search, view, and interact with data stored in Elasticsearch indexes, and intuitively presents data through various charts, tables, maps, and other formats, enabling advanced data analysis and visualization. It allows users to perform real-time log data analysis, system monitoring, and data visualization.

[0171] Flink is a framework and distributed processing engine for performing stateful computations on bounded and unbounded data streams. Designed for processing large-scale data streams, it runs in all common cluster environments, delivering high-performance, low-latency, real-time data processing at any scale.

[0172] First, the hyperconvergent computing monitoring system 100 collects key performance indicators of the devices in the hybrid architecture through the collection interface, such as the single-bit gate fidelity, single-bit coherence time, two-bit gate fidelity, single-bit read fidelity and overall read fidelity of the quantum computer. At the same time, the operation logs of the devices are also collected. Some implementation methods are: the hyperconvergent computing monitoring system 100 collects data from different sources in a distributed manner through Kafka, and sets multiple Kafka topics for different data sources. For example, the performance indicators and operation logs of different devices enter different Topics respectively, that is, the performance indicators and operation logs of device A enter Topic-A, and the performance indicators and operation logs of device B enter Topic-B.

[0173] Next, the hyperconverged computing monitoring system 100 obtains the target operating data of the target computing device according to the target subscription identifier. In some implementations, the hyperconverged computing monitoring system 100 uses Logstash to pull the target operating data of the target computing device from Kafka.

[0174] Then, the hyperconvergent computing monitoring system 100 preprocesses the target operation data to obtain the target data. Some implementation methods: For the performance indicators of the quantum computer, the hyperconvergent computing monitoring system 100 extracts key data such as quantum bit state information based on the custom parser in Logstash. For the operation log of the quantum computer, the hyperconvergent computing monitoring system 100 processes the operation log according to Logstash. The first step is to read the content of the log file. The second step is to extract key information such as timestamp, log level, message content, source IP address, etc. according to the structure and rules of the log, that is, to parse the log file. The third step is to convert the extracted information into a format that is more suitable for analysis. For example, the timestamp may need to be converted into a standard time format, and the IP address may need to be converted into geographic location information. The fourth step is to filter and screen the parsed data as needed to retain only the part of interest.

[0175] The processed data will be pushed to ElasticSearch and Oss for storage. In addition, ElasticSearch will also create different indexes for the processed data based on its own indexing function to ensure the efficiency of users when querying the operation log.

[0176] The hyperconverged computing monitoring system 100 then analyzes the newly acquired target data using Flink. It also uses Spark to pull historical operating data for offline analysis and predict the characteristic data of the target computing device. This helps maintain the stability and reliability of the hyperconverged computing monitoring system 100 and ensures efficient system operation. This means it can identify long-term performance trends or potential failure points, providing a basis for operations personnel to optimize system operation.

[0177] Finally, operations and maintenance personnel can create custom dashboards using Kibana in the hyperconvergent computing monitoring system 100 to monitor the system's operating status in real time. The quantum computing component has a dedicated dashboard that displays the quantum computer's operational status. The hyperconvergent computing monitoring system 100 also has alarm rules that assess whether preset alarm thresholds have been reached based on collected quantum computing performance indicators. If the thresholds are within the thresholds, the system updates the latest performance data and displays it in the foreground. If the thresholds are exceeded, the system automatically initiates a calibration process. After calibration, the system rechecks whether all indicators have returned to within the threshold range. If the target computing device has recovered, the data is updated and displayed in the foreground. If not, a calibration work order is sent to notify the quantum computing science team or operations and maintenance personnel to perform manual adjustments.

[0178] 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.

[0179] 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 media, USB flash drives, removable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media.

[0180] In the description of this specification, the descriptions with reference to the terms "particularly", "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 expressions 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.

[0181] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising 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 be performed out of 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 those skilled in the art to which the embodiments of the present application belong.

[0182] 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 computing resource monitoring method adapted to an over-convergent computing monitoring system, characterized in that: The method comprises: Based on a preset interface, acquiring target operation data of a target computing device according to a target subscription identifier, wherein the target subscription identifier is a subscription identifier that is pre-subscribed and corresponds to the target operation data; Preprocessing the target operation data to obtain target data; Based on a preset alarm threshold, the first operating state of the target computing device is monitored according to the target data to monitor the computing power resources of the super-convergent computing monitoring system.

2. The method according to claim 1, characterized in that The super-convergent computing monitoring system includes a plurality of computing devices, and the method further includes: A subscription identifier associated with each of the computing devices is set, the subscription identifier including the target subscription identifier.

3. The method according to claim 1, characterized in that The method of obtaining target operation data of a target computing device based on a preset interface and according to a target subscription identifier includes: Based on the preset interface, the operation data of all computing devices in the hyper-convergent computing monitoring system are acquired and stored, wherein the operation data includes performance indicators and operation logs of each computing device; The target operating data is acquired from the operating data according to the target subscription identifier.

4. The method according to claim 1, characterized in that: The preprocessing of the target operation data to obtain target data includes: Analyzing and processing the target operation data to obtain target analytical data; Converting the target parsed data to obtain target format data; The target format data is filtered to obtain target data.

5. The method according to claim 4, characterized in that The method further comprises: The historical operation data of the target computing device is updated according to the target data to predict the characteristic data of the target computing device according to the historical operation data, wherein the characteristic data includes the performance trend and potential failure points of the target computing device.

6. The method according to claim 1, characterized in that The monitoring of the first operating state of the target computing device according to the target data based on the preset alarm threshold includes: When the target data meets the preset alarm threshold, confirming that the first operating state is normal; When the target data does not meet the preset alarm threshold, it is confirmed that the first operating state is abnormal.

7. The method according to claim 6, characterized in that The method further comprises: If the first operating state is normal, visualization data is generated according to the target data, and the target data is displayed according to the visualization data.

8. The method according to claim 6, characterized in that The method further comprises: If the first operating state is abnormal, calibrating the target computing device within a preset calibration number of times to obtain resource calibration data; Based on the preset alarm threshold, the second operating state of the target computing device is monitored according to the resource calibration data to determine the operating state of the target computing device after the calibration process is performed.

9. The method according to claim 8, characterized in that The monitoring of the second operating state data of the target computing device according to the resource calibration data based on the preset alarm threshold and the preset calibration times includes: If the target computing device operates abnormally after the calibration process within the preset calibration times, a calibration work order is generated, and the calibration work order is used to indicate the abnormal operating state of the target computing device.

10. A super-convergent computing monitoring system, characterized in that: The super-convergent computing monitoring system includes a data collection module and a data analysis module: The data collection module is configured to obtain target operation data of a target computing device according to a target subscription identifier based on a preset interface, wherein the target subscription identifier is a subscription identifier that is pre-subscribed and corresponds to the operation data; The data analysis module is configured to pre-process the target operation data to obtain target data; The data analysis module is configured to monitor the first operating status data of the target computing device according to the target data based on a preset alarm threshold, so as to monitor the computing power resources of the super-convergent computing monitoring system.

11. An over-convergent computing monitoring server, characterized in that: The adapted hyperconvergent computing monitoring server includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method described in any one of claims 1 to 9 is implemented.

Images