Operation resource state monitoring management method and system

Abstract

The invention provides an operation resource state monitoring management method and system, and belongs to the technical field of state monitoring, and the method specifically comprises the steps: determining an optimization management information system in a problem information system based on the matching condition of historical problem processing data of a matching deviation problem type, according to real-time monitoring results of problem types of different reliable information systems and in combination with matching deviation problem types in different optimization management information systems, the management method for operation resources of large models of different optimization management information systems is determined, the efficiency and reliability of problem processing by using the large models are improved, and the problem processing efficiency is improved. And meanwhile, the abnormal occupation risk of multiple times of question and answer processing on computing power resources is also reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of status monitoring, and in particular relates to a method and system for monitoring and managing the status of operating resources. Background Art

[0002] By accessing the locally deployed large model, user issues can be automatically processed quickly and conveniently, thereby greatly reducing the processing pressure of information customer service. However, at the same time, how to monitor and process the operating resource status of the locally deployed large model to avoid technical problems such as low processing efficiency caused by data resource mismatch has become a technical problem that needs to be solved urgently.

[0003] To solve the above technical problems, the invention patent application CN202411941388.5, "A Resource Scheduling Method and Model Training Method for a Large Model," determines the deployment results of model instances deployed in multiple computing nodes through load prediction data, and performs resource scheduling based on the deployment results and current running resources. This solves the technical problem of low resource utilization of large models caused by low accuracy of resource scheduling of large models in related technologies. However, the above technical solution has the following technical problems: For large models, differences in training and processing data may lead to differences in the accuracy of their solution results. Therefore, users are more likely to repeatedly call large models to solve problems, resulting in a huge waste of operating resources. Therefore, how to monitor and manage the status of operating resources in a targeted manner and limit computing power resources in a targeted manner become technical issues that need to be solved urgently.

[0004] In order to solve the above technical problems, the present application provides a method and system for monitoring and managing the status of operating resources. Summary of the Invention

[0005] To achieve the purpose of the present invention, the present invention adopts the following technical solutions: Specifically, this application provides a method for monitoring and managing the status of running resources, which specifically includes: S1 determines the types of matching deviation problems in different information systems based on historical processing data. If it is determined that no unified supervision is required based on the distribution of matching deviation problem types in different information systems, then proceed to the next step. S2: dividing the information system into a problem information system and a reliable information system based on the distribution of the matching deviation problem type; determining, based on the historical problem processing data of the reliable information system, that it is necessary to monitor and manage the operating resources of the problem information system; and determining, based on the matching of the historical problem processing data of the matching deviation problem type, an optimized management information system in the problem information system; S3 determines the management method of the running resources of the large model of different optimization management information systems according to the real-time monitoring results of different reliable information system problem types and in combination with the matching deviation problem types in different optimization management information systems.

[0006] The present application has the following beneficial effects: Based on the matching of historical problem handling data of the matching deviation problem type, the optimization management information system in the problem information system is determined, so that the matching of historical problem handling data of the matching deviation problem type in the problem information system is realized, the determination of the change of training data caused by the update of problem handling data of different matching deviation problem types is realized, the screening of the optimization management system whose increased data volume does not meet the requirements is realized, and the technical problem of insufficient update reliability of historical problem handling data of the matching deviation problem type caused by the limitation of the background computing resource of the optimization management information system is avoided.

[0007] According to the real-time monitoring results of different reliable information system problem types and the matching deviation problem types in different optimization management information systems, the management method of the running resources of the large model of different optimization management information systems is determined, which not only considers the difference in the number of matching deviation problem types in different optimization management information systems, which leads to the difference in the potential occupation risk of overall computing resources, but also further combines the real-time monitoring results of reliable information system problem types to realize the determination of the computing resource limitation processing strategy of different optimization correlation information systems according to the difference in the real-time processing data of matching deviation problem types in reliable information systems, which not only ensures the reliability of problem handling of the optimization correlation information system, but also reduces the impact on the computing resources of the reliable information system.

[0008] The further technical solution is that the number of times of calling the large model required to obtain user satisfaction results is determined when the large model is used for processing, and it can be understood that when the number of times of calling the large model required to obtain user satisfaction results is more than 3 times, the problem type is determined as a matching deviation problem type.

[0009] Further, when the background running resources of the large model meet the requirements, there is no need to determine the matching deviation problem type, and the background running resources of the large model can be directly determined without monitoring management, that is, the background running resources of the large model are called according to the requirements of different information systems, and there is no need for background running resource limitation processing. In a possible embodiment, when the background running resources, that is, the servers in different time periods in the history are in an idle state, that is, there are idle computing nodes, it is determined that the background running resources of the large model meet the requirements Further technical solutions are to determine whether uniform supervision processing is required, specifically including: determining the number of matching deviation problem types in different information systems based on the composition data of the matching deviation problem types in different information systems; determining the total number of matching deviation problem types according to the number of matching deviation problem types in different information systems; determining whether uniform supervision processing is required based on the total number.

[0010] Further technical solutions are to determine the management method of the running resources of the large model of the optimization management information system, which is: determining the number of matching deviation problem types in different optimization management systems based on the historical problem processing data of the matching deviation problem types in different optimization management systems; determining the total number of problems according to the sum of the number of matching deviation problem types in the reliable information system and the number of matching deviation problem types in different optimization management systems; determining the management method of the running resources of the large model of the optimization management information system under the real-time monitoring result of the problem types in different reliable information systems according to the total number of problems and the number of matching deviation problem types in different optimization management systems.

[0011] In a second aspect, the present application provides a computer system, comprising a memory and a processor connected in communication, and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to perform the above-mentioned running resource state monitoring management method.

[0012] Other features and advantages will be set forth in the following description of the application, and in part will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the application. The objects and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims.

[0013] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are specifically described in detail below, and the accompanying drawings are referred to. BRIEF DESCRIPTION OF DRAWINGS

[0014] The above-mentioned and other features and advantages of the present application will become more apparent by describing in detail example embodiments thereof with reference to the attached drawings.

[0015] Figure 1 is a flowchart of a running resource state monitoring management method; Figure 2 is a flowchart of a method for determining whether uniform supervision processing is required; Figure 3 is a flowchart of a determined method of an optimization management information system in a problem information system; Figure 4 is a flowchart of a determined method of a management method of a running resource of a large model of an optimization management information system. DETAILED DESCRIPTION

[0016] Example embodiments now will be described more fully hereinafter with reference to the accompanying drawings. Example embodiments, however, can be implemented in many different forms and should not be construed as limited to the implementations set forth herein; rather, these implementations are provided as non-limiting examples so that this disclosure will be thorough and complete, and will fully convey the scope of the example embodiments to those skilled in the art. Like reference numerals refer to like elements throughout the figures, and thus description of the same will be simplified or omitted.

[0017] The terms "one", "a", "an", "the", and "said" are used to mean one or more elements, components, members, etc.; the terms "comprises", "comprising", "has", "having", "includes", "including", "contains", "containing", and the like are used to mean including, but not limited to.

[0018] In the present application, for the identification of the problem type that is difficult to solve in the large model, that is, the matching deviation problem type in the constitutive data of different information systems, and then generating differentiated computing resource limiting processing measures, thereby avoiding multiple calls to the large model processing, abnormal occupation of computing resources, and improving the efficiency of data processing.

[0019] Embodiment 1 To solve the above problems, according to one aspect of the present application, as shown in Figure 1 A running resource state monitoring management method is provided, specifically comprising: S1, based on historical processing data, determine the matching deviation problem type in different information systems, and according to the distribution of the matching deviation problem type in different information systems, determine whether to enter the next step when unified supervision processing is not required; Further, the matching deviation problem type is determined by the number of times of calling the large model required to obtain user satisfaction results when using the large model for processing. It can be understood that when the number of times of calling the large model required to obtain user satisfaction results is more than 3 times, the problem type is determined to be a matching deviation problem type.

[0020] Further, when the background running resources of the large model meet the requirements, it is not necessary to determine the matching deviation problem type, and it can be directly determined that monitoring management is not necessary, that is, the background running resources of the large model are called according to the requirements of different information systems, and the background running resources do not need to be limited. In one possible embodiment, when the background running resources, that is, the servers in different time periods in the history are all in an idle state, that is, there are idle computing nodes, it is determined that the background running resources of the large model meet the requirements Specifically, as shown in Figure 2 It is determined that unified supervision processing is not necessary, and specifically includes: The number of matching deviation problem types in different information systems is determined based on the constituent data of the matching deviation problem types in different information systems. The total number of matching deviation problem types is determined according to the number of matching deviation problem types in different information systems. Based on the total number, it is determined whether unified supervision processing is needed.

[0021] It can be understood that when the total number does not meet the requirements, that is, in the case of a large total number, if unified monitoring management of different information systems cannot be performed, the background computing power resources of the information system may meet the requirements. In one possible embodiment, when the total number is greater than 300 or more, it is determined that the total number does not meet the requirements.

[0022] It should be noted that the information system includes a data communication system, a voice switching system, a satellite communication system, a video conference system, and a power grid monitoring system. The customer service platform is constructed to realize problem processing of users in different information systems.

[0023] It should be noted that the matching deviation problem type is determined according to the analysis result of the key words of the problem type. Specifically, the same problem type is used for the same key word, and the same problem type can also be used for the same number of key words that meet the requirements.

[0024] It can be understood that when statistical supervision processing is needed, as long as there is a matching deviation problem type, that is, the key words of the problem type of the user of the information system are the same as the key words of the matching deviation problem type, the computing power resources of the information system are monitored and controlled, that is, the number of computing nodes is set to a preset number. In one possible embodiment, the preset number is the ratio of the number of all computing nodes to the number of information systems multiplied by 1 / 2.

[0025] It can be understood that it is determined that unified supervision processing is not necessary, and specifically includes: S11 determines the number of matching deviation problem types in different information systems based on the constituting data of the matching deviation problem types in different information systems, determines the total number of matching deviation problem types according to the number of matching deviation problem types in different information systems, and determines whether the total number of matching deviation problem types meets the requirement. Optionally, in the above step, if the total number does not meet the requirement, that is, in the case of a large total number, unified supervision processing is required.

[0026] In another embodiment, when the total number meets the requirement, it is further determined whether the number of information systems with matching deviation problem types meets the requirement. It can be understood that, when the number of information systems with matching deviation problem types is large, in a possible embodiment, all information systems have matching deviation problem types, and it is determined that unified supervision processing is required. When the number of information systems with matching deviation problem types meets the requirement, the next step is entered.

[0027] S12 determines the problem information system in the information system according to the number of matching deviation problem types in different information systems. It should be noted that, in a possible embodiment, the information system in which the information system has matching deviation problem types in different dates and the number of problems of the matching deviation problem types in different dates accounts for more than a preset proportion is regarded as a problem information system. Specifically, when it is more than 0.2, it is determined to be a problem information system.

[0028] It can be understood that, in the above step, if there is no problem information system, that is, there is no information system with a high degree of impact on the overall computing power resources, unified supervision processing is not required.

[0029] In addition, it should be further explained that, when there is a problem information system, it is further determined whether the number of problem information systems meets the requirement. When the number of problem information systems is large, in a possible embodiment, it is more than 2, it is determined that unified supervision processing is required.

[0030] In addition, it can be understood that, in the above step, if the number of problem information systems is not large, the next step is directly entered.

[0031] S13 determines whether unified supervision processing is required based on the total number and the number of problem information systems.

[0032] It can be understood that, in the above step, the total number and the number of problem information systems are used as the basis. Based on the total number, the number threshold of problem information systems is determined. When the number of problem information systems is greater than the number threshold, it is determined that statistical supervision processing is required.

[0033] It should be noted that the number threshold of the problem information system can be determined according to the ratio of the preset number to the total number.

[0034] S2 divides the information system into a problem information system and a reliable information system according to the distribution of the matching deviation problem type, and determines the optimal management information system in the problem information system based on the matching of the historical problem processing data of the matching deviation problem type, according to the historical problem processing data of the reliable information system, when the monitoring and management processing of the running resources of the problem information system is required. Further, the information system is divided into a problem information system and a reliable information system according to the distribution of the matching deviation problem type, and specifically includes: The information system with the number of matching deviation problem types meeting the requirements is regarded as a problem information system, and the others are regarded as reliable information systems.

[0035] It can be understood that the information system has matching deviation problem types in different dates, and the problem number ratio of the matching deviation problem types in different dates is greater than the preset ratio. Specifically, it is determined that the monitoring and management processing of the running resources of the problem information system is required, and specifically includes: According to the historical problem processing data of the reliable information system, the computing nodes required for processing only the problems of the reliable information system are determined and used as the used computing nodes. According to the difference between the number of all computing nodes and the number of used computing nodes, the number of available computing nodes is determined. The number of available computing nodes in different time periods is determined to determine whether the monitoring and management processing of the running resources of the problem information system is required.

[0036] It can be understood that when the number of available computing nodes does not meet the requirement of the number of time periods, it is determined that the monitoring and management processing of the running resources of the problem information system is required, and when the monitoring and management processing of the running resources of the problem information system is not required, it can be directly determined that the monitoring and management is not required, that is, the background running resources of the large model are called according to the requirements of different information systems, and the limitation processing of the background running resources is not required.

[0037] Specifically, in one possible embodiment, when the ratio of the available computing nodes to all computing nodes is determined, the available node ratio is determined, and when the time period ratio of the available node ratio less than 0.2 is greater than 0.3, it is determined that the monitoring and management processing of the running resources of the problem information system is required.

[0038] In another possible embodiment, when the average of the ratio of the number of available computing nodes in different time periods to the number of problem information systems is less than a preset proportion threshold, it is determined that monitoring and management processing of the running resources of the problem information system needs to be performed, where the preset proportion threshold is determined according to the average of the number of computing nodes required by different problem information systems to process problems.

[0039] Specifically, as shown in Figure 3 The method for determining the optimization management information system in the problem information system is: Based on the matching condition of the historical problem processing data of the matching deviation problem type, the number of problem processing times of the matching deviation problem type in history is determined. Based on the matching condition of the historical problem processing data of the matching deviation problem type, the number of problem processing times of the matching deviation problem type in history is determined. Based on the number of problem processing times of different matching deviation problem types, it is determined whether the problem information system is an optimization management information system.

[0040] Specifically, the number of problem processing times is the number of processing times of the problem of the matching deviation problem type in history.

[0041] It should be noted that when the number of problem processing times of different matching deviation problem types in the problem information system is large, a large amount of training data has been accumulated, and on this basis, it is determined that it does not belong to the optimization management information system. It should be noted that the number of problem processing times can be determined by a fixed threshold, where when it is greater than a preset processing time threshold, it is determined that it is large. In a possible embodiment, the preset processing time threshold is determined according to the average number of interactive processing times of the problem of the matching deviation problem type. In a possible embodiment, it is set to 100 times or more, that is, the average number of interactive processing times between the closest training completion time is 100 times or more.

[0042] It can be understood that when the problem information system does not belong to the optimization management information system, the number of computing nodes of the problem information system is directly fixed to a fixed number. In a possible embodiment, the specific fixed number is determined according to the ratio of the number of all computing nodes to the number of information systems multiplied by 2 / 3, so as to provide more computing nodes for the optimization management information system.

[0043] It should be noted that the fixed number is determined according to the number of optimization management information systems, where the more the number of optimization management information systems, the smaller the fixed number.

[0044] S3 determines the management method of the running resource of the large model of the different optimization management information systems according to the real-time monitoring result of the different reliable information system problem types and in combination with the matching deviation problem types in the different optimization management information systems.

[0045] Specifically, as shown in Figure 4 The determined method of the management method of the running resource of the large model of the optimization management information system is: determining the number of the matching deviation problem types in the different optimization management systems according to the historical problem processing data of the matching deviation problem types in the different optimization management systems; determining the total number of problems according to the sum of the number of the matching deviation problem types in the reliable information system and the number of the matching deviation problem types in the different optimization management systems; determining the management method of the running resource of the large model of the optimization management information system under the real-time monitoring result of the different reliable information system problem types according to the total number of problems and the number of the matching deviation problem types in the different optimization management systems.

[0046] It should be noted that when the total number of problems is greater than the preset problem quantity threshold, for all optimization management systems, as long as the reliable information system has the matching deviation problem type in the current user's problem, that is, there is an unprocessed matching deviation problem type, all optimization management systems need to be subjected to the processing of limiting the computing resource, and in a possible embodiment, when the ratio of the total number of problems, the sum of the number of reliable information systems and the number of optimization management information systems is greater than 50 or more, it is determined that the total number of problems does not meet the requirement, and when there is a matching deviation problem type, the number of computing nodes of the optimization management information system is directly fixed at a fixed number.

[0047] When the total number of problems is not greater than the preset problem quantity threshold, the number of the matching deviation problem types in the different optimization management systems also needs to be determined at this time, and for the optimization management system with more matching deviation problem types, as long as the reliable information system has the matching deviation problem type in the current user's problem, the optimization management system with more matching deviation problem types needs to be subjected to the processing of limiting the computing resource, that is, the number of computing nodes of the optimization management information system is fixed at a fixed number, and in a possible embodiment, when the number of matching deviation problem types is 70 or more, it is determined that the number of matching deviation problem types is more.

[0048] For the optimization correlation system with less matching deviation problem type, when it is determined that there is a matching deviation problem type in the current user's problem of the reliable information system according to the real-time monitoring data of the reliable information system, the number of matching deviation problem types is obtained, and when the number of matching deviation problem types in the current user's problem of the reliable information system does not meet the requirement, the computing resource of the optimization correlation system with less matching deviation problem type is limited. It should be noted that in one possible embodiment, when the number of matching deviation problem types is not more than 70, when the problem quantity proportion of matching deviation problem types in the current user's problem of different reliable information systems is more than 0.3 and the proportion of the number of remaining available computing nodes is less than a threshold value, it is determined that the computing resource of the optimization correlation system with less matching deviation problem type is limited, wherein the threshold value is determined according to the number of all computing nodes, and the larger the number of all computing nodes, the smaller the threshold value. In one possible embodiment, it can be set to 0.05.

[0049] Optionally, the determined method of the management method of the running resource of the large model of the optimization management information system is: S41, determining the deviation influence value of different optimization management systems according to the number of matching deviation problem types in different optimization management systems; It should be noted that before determining the deviation influence value, it is also necessary to determine whether the total number of matching deviation problem types in different optimization management systems meets the requirement. When the total number of matching deviation problem types in different optimization management systems does not meet the requirement, i.e. the number is greater than the threshold value, it can be directly determined that as long as the reliable information system has a matching deviation problem type in the current user's problem, the optimization management system needs to be limited in computing resource. Optionally, when the total number of matching deviation problem types in different optimization management systems meets the requirement, in one possible embodiment, the deviation influence value is determined according to the number of matching deviation problem types in the above step, specifically according to the ratio of the number of matching deviation problem types to a preset fixed threshold value, wherein the preset fixed threshold value is determined according to the number of reliable management systems, and the more the number of reliable management systems, the larger the preset fixed threshold value, and in one possible embodiment, the value is 100.

[0050] It can be understood that in the above steps, if the number of matching deviation problem types is large, that is, the deviation influence value is large, that is, greater than the preset threshold, then due to the large risk of occupying computing resources caused by matching deviation problem types, on this basis, it can be directly determined that as long as the reliable information system has matching deviation problem types in the current user's problem, the optimization management system needs to be limited in computing resources; In addition, it should be noted that if the number of matching deviation problem types is not large, that is, the deviation influence value is not large, it is further necessary to determine whether the sum of the deviation influence values of different optimization management systems meets the requirements. Specifically, when the sum of the deviation influence values of different optimization management systems is large, that is, greater than the threshold, it can be directly determined that as long as the reliable information system has matching deviation problem types in the current user's problem, the optimization management system needs to be limited in computing resources; Further, even if the sum of the deviation influence values of different optimization management systems meets the requirements, it is further necessary to turn to the next step; S42 uses the historical problem processing data of the matching deviation problem types in different optimization management systems to determine the number of matching deviation problem types in different optimization management systems, and determines the total number of problems according to the sum of the number of matching deviation problem types in the reliable information system and the number of matching deviation problem types in different optimization management systems; In addition, it should be noted that in one possible embodiment, when the total number of problems is greater than the preset problem number threshold, for all optimization management systems, as long as the reliable information system has matching deviation problem types in the current user's problem, that is, there are unprocessed matching deviation problem types, all optimization management systems need to be limited in computing resources. In one possible embodiment, when the ratio of the sum of the total number of problems, the number of reliable information systems and optimization management information systems is greater than 50 or more, it is determined that the total number of problems does not meet the requirements, and when there is a matching deviation problem type, the number of optimization management information system computing nodes is directly fixed at a fixed number.

[0051] Further, when the total number of problems is not greater than the preset problem number threshold, it is further necessary to determine that when the total number of problems is small, that is, not in the preset interval, whether the reliable information system has a matching deviation problem type or not, it is determined that the number of optimization management information system computing nodes does not need to be limited.

[0052] When the total number of problems is not small, i.e., in the preset interval, if the proportion of the number of matching deviation problem types in the different optimization management systems in the total number of problems is large, i.e., greater than the preset proportion threshold, when the matching deviation problem type exists, the number of computing nodes of the optimization management information system is directly fixed to a fixed number.

[0053] S43, according to the total number of problems and the deviation influence values in the different optimization management systems, determines a management method of the running resources of the large model of the optimization management information system under the real-time monitoring result of the problem type of the different reliable information systems.

[0054] Further, in the above steps, the number of matching deviation problem types in the different optimization management systems also needs to be determined. For the optimization management system with more matching deviation problem types, as long as the reliable information system has the matching deviation problem type in the current user's problem, the optimization management system with more matching deviation problem types needs to be subjected to the computing resource limiting processing, i.e., the number of computing nodes of the optimization management information system is fixed to a fixed number. In a possible embodiment, when the number of matching deviation problem types is more than 70, it is determined that the number of matching deviation problem types is more.

[0055] For the optimization management system with less matching deviation problem types, according to the real-time monitoring data of the reliable information system, when the reliable information system has the matching deviation problem type in the current user's problem, the number of matching deviation problem types is obtained. When the number of matching deviation problem types in the current user's problem in the reliable information system does not meet the requirement, the optimization management system with less matching deviation problem types is subjected to the computing resource limiting processing. It should be noted that, in a possible embodiment, when the number of matching deviation problem types is not more than 70, when the proportion of the number of matching deviation problem types in the current user's problem in the different reliable information systems is more than 0.3 and the proportion of the number of remaining available computing nodes is less than a threshold, the optimization management system with less matching deviation problem types is subjected to the computing resource limiting processing, wherein the threshold is determined according to the number of all computing nodes, and the larger the number of all computing nodes is, the smaller the threshold is. In a possible embodiment, the threshold can be set to 0.05.

[0056] Embodiment 2 In a second aspect, the present application provides a computer system, comprising a memory and a processor connected in communication, and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to perform the above-mentioned running resource state monitoring management method.

[0057] The various embodiments in this specification describe the application in progressive stages. Identical or similar parts from one embodiment to another embodiment can be mutually referred to. Each embodiment focuses on the difference from other embodiments. In particular, the device, apparatus, and non-transitory computer storage medium embodiments are described more simply because they are substantially similar to the method embodiments. The relevant parts can be referred to the method embodiment description.

[0058] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims can be performed in a different order than the order in which they are recited and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown or sequential order in order to achieve the desired results. In some implementations, multitasking and parallel processing can be advantageous or necessary.

[0059] The above only describes one or more embodiments of the present specification and is not intended to limit the present specification. One or more embodiments of the present specification can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of one or more embodiments of the present specification shall be included in the scope of the claims of the present specification.

Claims

1. A method for monitoring and managing the status of operating resources, characterized in that: Specifically include: Based on historical processing data, determine the types of matching deviation issues in different information systems. If it is determined that no unified regulatory processing is required based on the distribution of matching deviation issue types in different information systems, proceed to the next step. The information system is divided into a problem information system and a reliable information system according to the distribution of the matching deviation problem type; when it is determined that the operation resources of the problem information system need to be monitored and managed based on the historical problem processing data of the reliable information system, an optimized management information system in the problem information system is determined based on the matching of the historical problem processing data of the matching deviation problem type; According to the real-time monitoring results of the problem types of different reliable information systems and combined with the matching deviation problem types in different optimization management information systems, the management methods of the operating resources of the large models of different optimization management information systems are determined.

2. The method for monitoring and managing the operating resource status according to claim 1, wherein: The matching deviation problem type is determined by the number of times the large model is called to obtain a user-satisfied result when the large model is used for processing.

3. The method for monitoring and managing the operating resource status according to claim 1, wherein: When the background running resources of the large model meet the requirements, it is determined that no monitoring and management is required.

4. The method for monitoring and managing the operating resource status according to claim 1, wherein: It is determined that no unified regulatory treatment is required, including: Determine the number of types of matching deviation problems in different information systems based on constituent data of the types of matching deviation problems in different information systems; Determine the total number of matching deviation problem types based on the number of matching deviation problem types in different information systems; Based on the total quantity, a determination is made as to whether unified regulatory treatment is required.

5. The method for monitoring and managing the operating resource status according to claim 1, wherein: The information system includes a data communication system, a voice exchange system, a satellite communication system, a video and telephone conference system, and a power grid monitoring system.

6. The method for monitoring and managing the operating resource status according to claim 1, wherein: The information system is divided into a problem information system and a reliable information system according to the distribution of the matching deviation problem types, specifically including: The information systems whose number of matching deviation problem types meets the requirements are regarded as problematic information systems, and the others are regarded as reliable information systems.

7. The method for monitoring and managing the operating resource status according to claim 1, wherein: The method for determining the optimized management information system in the problem information system is: Based on the matching of the historical problem processing data of the matching deviation problem type, determine the optimization management information system in the problem information system Determine the number of problem processing times of the matching deviation problem type in history based on the matching status of the historical problem processing data of the matching deviation problem type; Based on the number of problem processing times of different matching deviation problem types, it is determined whether the problem information system is an optimization management information system.

8. The method for monitoring and managing the operating resource status according to claim 7, wherein: When the problem information system does not belong to the optimization management information system, the number of its computing power nodes is directly fixed at a fixed number.

9. The method for monitoring and managing the operating resource status according to claim 1, wherein: The method for determining the management method of the operating resources of the large model of the optimization management information system is: Determine the number of matching deviation problem types in different optimization management systems using historical problem processing data of matching deviation problem types in different optimization management systems; determining a total number of problems based on the sum of the number of matching deviation problem types in the reliable information system and the number of matching deviation problem types in different optimization management systems; According to the total number of problems and the number of matching deviation problem types in different optimization management systems, a management method for optimizing the operating resources of the large model of the management information system is determined under the real-time monitoring results of the problem types of different reliable information systems.

10. A computer system comprising: A memory and a processor in communication connection, and a computer program stored in the memory and capable of running on the processor, characterized in that when the processor runs the computer program, it executes a method for monitoring and managing the status of running resources as described in any one of claims 1-9.

Images