Cloud resource configuration method and system based on classification of multiple private cloud management domains

Abstract

The invention provides a cloud resource configuration method and system based on classification of multiple private cloud management domains, and the method comprises the steps: obtaining the cloud resource configuration historical data of all private cloud management domains from the cloud resource configuration historical records of the private cloud management domains, and carrying out the preprocessing and data correction processing of the cloud resource configuration historical data, obtaining effective cloud resource configuration record data; carrying out clustering analysis processing on the effective cloud resource configuration record data, and carrying out identification analysis on similarity and difference among different management domains based on respective cloud resource configuration feature information of all private cloud management domains so as to divide all private cloud management domains into a plurality of management domain groups; the management domain group is conveniently utilized to provide cloud resource configuration in a centralized and unified manner subsequently; the cloud resource demand attribute information of the new nanotube management domain is determined based on the work log of the new nanotube management domain, the new nanotube management domain is identified and classified by using the management domain group so as to select the management domain group matched with the new nanotube management domain, and the new nanotube management domain is identified and classified based on the region and the service type of the new nanotube management domain. And performing accurate and efficient cloud resource demand prediction according to the existing data of the new nano management domain and the historical data of the matched management domain group.

Description

Technical Field

[0001] The present invention relates to the field of cloud resources, and in particular to a cloud resource configuration method and system based on multi-private cloud management domain classification. Background Art

[0002] As cloud service providers continue to expand the scale of cloud computing resources, scientifically and effectively predicting cloud resource demand is essential for better planning and management of cloud computing resources. In order to accurately and efficiently predict cloud resource demand in a private cloud environment, it is an important topic to classify and analyze different management domains with different cloud resource configuration data. However, the existing methods for predicting private cloud resource demand do not accurately identify and analyze the similarities and differences between different management domains, and cannot fully understand the characteristics and trends of different management domains. They cannot fully utilize the most matching management domain cloud resource historical data to timely and accurately predict the cloud resource demand of newly managed management domains with less existing data. This easily leads to low cloud resource demand prediction accuracy and erroneous predictions, and cannot fully and accurately utilize each management domain for reliable cloud resource configuration. Summary of the invention

[0003] The purpose of the present invention is to provide a cloud resource configuration method and system based on the classification of multiple private cloud management domains, which obtains the cloud resource configuration history data of all private cloud management domains from the cloud resource configuration history records of several private cloud management domains, and pre-processes and corrects the cloud resource configuration history data to obtain effective cloud resource configuration record data; then cluster analysis is performed on the effective cloud resource configuration record data, and based on the cloud resource configuration feature information of all private cloud management domains, the similarities and differences between different management domains are accurately identified and analyzed, so as to divide all private cloud management domains into several management domain groups, so as to facilitate the subsequent use of management domain groups to provide cloud resource configuration in a centralized and unified manner; based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the management domain group is used to identify and classify the newly managed management domain, so as to select the management domain group that matches the newly managed management domain, and based on the location and business type of the newly managed management domain, accurate and efficient cloud resource demand prediction is performed according to the existing data of the newly managed management domain and the historical data of the matched management domain group.

[0004] The present invention is achieved through the following technical solutions: The cloud resource configuration method based on multi-private cloud management domain classification includes: Based on the cloud resource configuration history records of each of the several private cloud management domains, the cloud resource configuration history data of each of the private cloud management domains is obtained, and the cloud resource configuration history data is preprocessed; data correction processing is performed on the cloud resource configuration history data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data; Performing cluster analysis on the effective cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains; based on the cloud resource configuration feature information, identifying and analyzing the similarities and differences between different management domains, and dividing all private cloud management domains into a number of management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions; Based on the work log of the newly managed management domain, determine the cloud resource demand attribute information of the newly managed management domain, use the management domain group to identify and classify the newly managed management domain, so as to select the management domain group that matches the newly managed management domain; then, based on the location and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matching management domain group, perform cloud resource demand prediction.

[0005] Optionally, based on the cloud resource configuration history records of each of the several private cloud management domains, the cloud resource configuration history data of each of the private cloud management domains is obtained, and the cloud resource configuration history data is preprocessed; and data correction processing is performed on the cloud resource configuration history data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data, including: Obtain cloud resource configuration history records of several private cloud management domains in a preset historical time interval, analyze the cloud resource configuration history records, and obtain all cloud resource configuration valid behaviors that occurred in each private cloud management domain in the preset historical time interval; analyze all cloud resource configuration valid behaviors to determine the configuration history data of each private cloud management domain on different types of cloud resources; wherein the configuration history data at least includes the configuration occurrence time and the number of configured resources of the corresponding type of cloud resources; the different types of cloud resources include computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources; and perform noise removal preprocessing and missing data completion preprocessing on the configuration history data in turn; Based on the Holt-Winters model, data smoothing processing is performed on the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing; then based on the result of the data smoothing processing, deviation and abnormal data component elimination and correction processing is performed on the configuration history data, so as to obtain effective cloud resource configuration record data.

[0006] Optionally, cluster analysis is performed on the valid cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains; based on the cloud resource configuration feature information, similarities and differences between different management domains are identified and analyzed, and all private cloud management domains are divided into several management domain groups; wherein all private cloud management domains under each management domain group meet corresponding cloud resource configuration and change trend similarity conditions, including: Based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources; Based on the demand change trend characteristic information, determine the cloud resource type information matched and configured by all private cloud management domains; then based on the matched and configured cloud resource type information, divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group are matched and configured with the same type of cloud resources, and the demand change trends of the cloud resources matched and configured by all private cloud management domains under each management domain group are within a preset quantity range.

[0007] Optionally, based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the newly managed management domain is identified and classified using the management domain group, so as to select a management domain group that matches the newly managed management domain; and based on the region and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matching management domain group, cloud resource demand prediction is performed, including: Analyze the work log of the newly managed management domain to obtain the expected configuration cloud resource type information of the computing task currently executed by the newly managed management domain; compare the expected configuration cloud resource type information with the cloud resource type information that all management domain groups can provide for configuration, and select the management domain group that matches the newly managed management domain; then, based on the location and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matching management domain group, predict the cloud resource demand.

[0008] Optionally, the operation quality of the acquisition process of the work log of the newly managed management domain is determined, including: Extract the acquisition time of the work log; Compare the acquisition time of the work log with a preset acquisition time threshold; When the acquisition time of the work log exceeds the preset acquisition time threshold, the operation parameters for acquiring the work log are retrieved, wherein the operation parameters include the diary screening accuracy, the diary retrieval garbled code ratio and the diary retrieval response time; The diary screening accuracy, diary retrieval garbled code ratio and diary retrieval response time are used to obtain a diary retrieval operation evaluation coefficient; The diary retrieval operation evaluation coefficient is obtained by the following formula: Where W represents the diary retrieval operation evaluation coefficient; n represents the number of diary retrieval times; P irepresents the diary screening accuracy corresponding to the i-th diary retrieval; L i represents the proportion of garbled characters in the diary retrieval corresponding to the i-th diary retrieval; T i T represents the diary retrieval response time corresponding to the i-th diary retrieval; Lmax Indicates the diary retrieval response time corresponding to the maximum value of the diary retrieval garbled code ratio; T Lmin It indicates the diary retrieval response time corresponding to the minimum value of the diary retrieval garbled code ratio; S indicates the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Among them, S represents the adjustment coefficient; L pmin Indicates the proportion of garbled characters in diary retrieval corresponding to the minimum diary screening accuracy; L pmax Indicates the proportion of garbled characters in diary retrieval corresponding to the maximum diary screening accuracy; L min Indicates the minimum value of the garbled character ratio in diary retrieval; L max Indicates the maximum ratio of garbled characters retrieved from the diary; Comparing the diary retrieval operation evaluation coefficient with a preset coefficient threshold; When the diary retrieval operation evaluation coefficient is lower than a preset coefficient threshold, it is determined that there is an abnormality in the work log acquisition process, and an abnormality alarm is issued.

[0009] The cloud resource configuration system based on multi-private cloud management domain classification includes: A historical data acquisition and preprocessing module, used to acquire the cloud resource configuration historical data of all private cloud management domains based on the cloud resource configuration historical records of the respective private cloud management domains, and preprocess the cloud resource configuration historical data; A historical data correction module, used to perform data correction processing on the cloud resource configuration historical data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data; A cloud resource configuration feature identification module is used to perform cluster analysis on the valid cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains; A management domain group division module is used to identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions; A management domain group selection module is used to determine the cloud resource demand attribute information of the newly managed management domain based on the work log of the newly managed management domain, identify and classify the newly managed management domain using the management domain group, and select a management domain group that matches the newly managed management domain; The cloud resource configuration prediction module is used to predict cloud resource demand based on the region and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matched management domain group.

[0010] Optionally, the historical data acquisition and preprocessing module is used to acquire the cloud resource configuration historical data of all private cloud management domains based on the cloud resource configuration history records of the respective private cloud management domains, and preprocess the cloud resource configuration historical data, including: Obtain cloud resource configuration history records of several private cloud management domains in a preset historical time interval, analyze the cloud resource configuration history records, and obtain all cloud resource configuration valid behaviors that occurred in each private cloud management domain in the preset historical time interval; analyze all cloud resource configuration valid behaviors to determine the configuration history data of each private cloud management domain on different types of cloud resources; wherein the configuration history data at least includes the configuration occurrence time and the number of configured resources of the corresponding type of cloud resources; the different types of cloud resources include computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources; and perform noise removal preprocessing and missing data completion preprocessing on the configuration history data in turn; The historical data correction module is used to perform data correction processing on the cloud resource configuration historical data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data, including: Based on the Holt-Winters model, data smoothing processing is performed on the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing; then based on the result of the data smoothing processing, deviation and abnormal data component elimination and correction processing is performed on the configuration history data, so as to obtain effective cloud resource configuration record data.

[0011] Optionally, the cloud resource configuration feature identification module is used to perform cluster analysis on the valid cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains, including: Based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources; The management domain group division module is used to identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions, including: Based on the demand change trend characteristic information, determine the cloud resource type information matched and configured by all private cloud management domains; then based on the matched and configured cloud resource type information, divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group are matched and configured with the same type of cloud resources, and the demand change trends of the cloud resources matched and configured by all private cloud management domains under each management domain group are within a preset quantity range.

[0012] Optionally, the management domain group selection module is used to determine the cloud resource demand attribute information of the newly managed management domain based on the work log of the newly managed management domain, and use the management domain group to identify and classify the newly managed management domain, so as to select a management domain group matching the newly managed management domain, including: Analyze the work log of the newly managed management domain to obtain the expected configuration cloud resource type information of the computing task currently executed by the newly managed management domain; compare the expected configuration cloud resource type information with the cloud resource type information that all management domain groups can provide for configuration, and select the management domain group that matches the newly managed management domain.

[0013] Optionally, the operation quality of the acquisition process of the work log of the newly managed management domain is determined, including: Extract the acquisition time of the work log; Compare the acquisition time of the work log with a preset acquisition time threshold; When the acquisition time of the work log exceeds the preset acquisition time threshold, the operation parameters for acquiring the work log are retrieved, wherein the operation parameters include the diary screening accuracy, the diary retrieval garbled code ratio and the diary retrieval response time; The diary screening accuracy, diary retrieval garbled code ratio and diary retrieval response time are used to obtain a diary retrieval operation evaluation coefficient; The diary retrieval operation evaluation coefficient is obtained by the following formula: Where W represents the diary retrieval operation evaluation coefficient; n represents the number of diary retrieval times; P i represents the diary screening accuracy corresponding to the i-th diary retrieval; L i represents the proportion of garbled characters in the diary retrieval corresponding to the i-th diary retrieval; T i T represents the diary retrieval response time corresponding to the i-th diary retrieval; Lmax Indicates the diary retrieval response time corresponding to the maximum value of the diary retrieval garbled code ratio; T Lmin It indicates the diary retrieval response time corresponding to the minimum value of the diary retrieval garbled code ratio; S indicates the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Among them, S represents the adjustment coefficient; L pmin Indicates the proportion of garbled characters in diary retrieval corresponding to the minimum diary screening accuracy; L pmax Indicates the proportion of garbled characters in diary retrieval corresponding to the maximum diary screening accuracy; L min Indicates the minimum value of the garbled character ratio in diary retrieval; L max Indicates the maximum ratio of garbled characters retrieved from the diary; Comparing the diary retrieval operation evaluation coefficient with a preset coefficient threshold; When the diary retrieval operation evaluation coefficient is lower than a preset coefficient threshold, it is determined that there is an abnormality in the work log acquisition process, and an abnormality alarm is issued.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The cloud resource configuration method and system based on multi-private cloud management domain classification provided in the present application obtain the cloud resource configuration historical data of all private cloud management domains from the cloud resource configuration historical records of several private cloud management domains, and perform preprocessing and data correction processing on the cloud resource configuration historical data to obtain effective cloud resource configuration record data; then perform cluster analysis processing on the effective cloud resource configuration record data, and accurately identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information of all private cloud management domains, so as to divide all private cloud management domains into several management domain groups, so as to facilitate the subsequent use of management domain groups to provide cloud resource configuration in a centralized and unified manner; further, based on the work log of the newly managed management domain, determine the cloud resource demand attribute information of the newly managed management domain, use the management domain group to identify and classify the newly managed management domain, so as to select the management domain group that matches the newly managed management domain, and based on the location and business type of the newly managed management domain, perform accurate and efficient cloud resource demand prediction based on the existing data of the newly managed management domain and the historical data of the matched management domain group. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them: Figure 1 A schematic diagram of the flow of a cloud resource configuration method based on multi-private cloud management domain classification provided by the present invention.

[0016] Figure 2 A schematic diagram of the structure of a cloud resource configuration system based on multi-private cloud management domain classification provided by the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0018] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0019] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0020] See also Figure 1 As shown, an embodiment of the present application provides a cloud resource configuration method based on multiple private cloud management domain classification. The cloud resource configuration method based on multiple private cloud management domain classification includes: Based on the cloud resource configuration history records of several private cloud management domains, the cloud resource configuration history data of all private cloud management domains are obtained, and the cloud resource configuration history data is preprocessed; data correction processing is performed on the cloud resource configuration history data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data; Perform cluster analysis on the effective cloud resource configuration record data to obtain the cloud resource configuration feature information of all private cloud management domains; based on the cloud resource configuration feature information, identify and analyze the similarities and differences between different management domains, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions; Based on the work log of the newly managed management domain, determine the cloud resource demand attribute information of the newly managed management domain, use the management domain group to identify and classify the new management domain, and select the management domain group that matches the new management domain; then, based on the location and business type of the newly managed management domain and the existing data of the new management domain and the historical data of the matching management domain group, perform cloud resource demand prediction.

[0021] The beneficial effects of the above embodiments are as follows: the cloud resource configuration method based on multi-private cloud management domain classification obtains the cloud resource configuration historical data of all private cloud management domains from the cloud resource configuration historical records of several private cloud management domains, and pre-processes and corrects the cloud resource configuration historical data to obtain effective cloud resource configuration record data; then cluster analysis is performed on the effective cloud resource configuration record data, and based on the cloud resource configuration feature information of all private cloud management domains, the similarities and differences between different management domains are accurately identified and analyzed, so as to divide all private cloud management domains into several management domain groups, so as to facilitate the subsequent use of management domain groups to provide cloud resource configuration in a centralized and unified manner; based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the management domain group is used to identify and classify the newly managed management domain, so as to select the management domain group that matches the newly managed management domain, and based on the location and business type of the newly managed management domain, accurate and efficient cloud resource demand prediction is performed according to the existing data of the newly managed management domain and the historical data of the matched management domain group.

[0022] In another embodiment, based on the cloud resource configuration history records of each of the several private cloud management domains, the cloud resource configuration history data of each of the private cloud management domains is obtained, and the cloud resource configuration history data is preprocessed; and the cloud resource configuration history data that has completed the preprocessing is subjected to data correction processing, thereby obtaining valid cloud resource configuration record data, including: Obtain the cloud resource configuration history records of several private cloud management domains in a preset historical time interval, analyze the cloud resource configuration history records, and obtain all cloud resource configuration valid behaviors that occurred in each private cloud management domain in the preset historical time interval; analyze all cloud resource configuration valid behaviors to determine the configuration history data of each private cloud management domain on different types of cloud resources; wherein the configuration history data at least includes the configuration occurrence time and the number of configured resources of the corresponding type of cloud resources; the different types of cloud resources include computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources; and perform noise removal preprocessing and missing data completion preprocessing on the configuration history data in turn; Based on the Holt-Winters model, data smoothing is performed on the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing. Based on the result of the data smoothing, deviation and abnormal data component elimination and correction processing is performed on the configuration history data to obtain effective cloud resource configuration record data.

[0023] The beneficial effect of the above embodiment is that different private cloud management domains can perform configuration operations of different types of cloud resources, and the different types of cloud resources may include but are not limited to computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources, etc. Different private cloud management domains usually tend to concentrate on configuring one type of cloud resources in actual cloud resource configuration operations. In order to accurately identify the cloud resource configuration characteristics of all private cloud management domains, the cloud resource configuration history records of several private cloud management domains in a preset historical time interval are obtained and analyzed to obtain all valid cloud resource configuration behaviors that occurred in each private cloud management domain in the preset historical time interval, wherein the valid cloud resource configuration behavior may be but is not limited to a behavior in which the cloud resource configuration duration is greater than a preset time length threshold; and the valid cloud resource configuration behavior is analyzed to determine the configuration history data of each private cloud management domain on different types of cloud resources, so that the cloud resource configuration tendency of each private cloud management domain can be quantitatively collected, providing sufficient and comprehensive data support for the subsequent similarity relationship analysis between different private cloud management domains. In addition, the configuration history data is subjected to noise removal preprocessing and missing data completion preprocessing in sequence, so as to remove the noise interference components in the configuration history data and improve the completeness of the data; and based on the Holt-Winters model, the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing is subjected to data smoothing processing; and based on the result of the data smoothing processing, the configuration history data is subjected to deviation and abnormal data component elimination and correction processing, so as to obtain effective cloud resource configuration record data, thereby improving data quality and accuracy, and enhancing the reliability and accuracy of subsequent similarity relationship identification between different private cloud management domains.

[0024] In another embodiment, cluster analysis is performed on the effective cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains; based on the cloud resource configuration feature information, similarities and differences between different management domains are identified and analyzed, and all private cloud management domains are divided into several management domain groups; wherein all private cloud management domains under each management domain group meet corresponding cloud resource configuration and change trend similarity conditions, including: Based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources; Based on the demand change trend characteristic information, determine the cloud resource type information that is matched and configured for all private cloud management domains; then, based on the cloud resource type information that is matched and configured, divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group are matched and configured with the same type of cloud resources, and the demand change trends of the cloud resources that are matched and configured for all private cloud management domains under each management domain group are within a preset quantity range.

[0025] The beneficial effect of the above embodiment is that each private cloud management domain can simultaneously manage computing power cloud resources, cloud memory resources, storage cloud resources, network bandwidth cloud resources, etc. However, in actual cloud resource configuration operations, each private cloud management domain does not necessarily configure all types of cloud resources managed by its subordinates equally, that is, each private cloud management domain may tend to mainly configure one or two of the cloud resources. At this level, all private cloud management domains can be distinguished by similarity. To this end, based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources. Based on the demand change trend characteristic information, the cloud resource type information of all private cloud management domains is determined, so that all private cloud management domains are divided into several management domain groups; among which, all private cloud management domains under each management domain group match and configure the same type of cloud resources, and the demand change trend of cloud resources matched and configured by all private cloud management domains under each management domain group is within the preset quantity range, so that all private cloud management domains under the same management domain group tend to concentrate on configuring the same type of cloud resources. Through cluster analysis based on the hybrid algorithm of DBSCAN and Kmeans, the similarities and differences between different private management domains can be accurately identified, providing more accurate management domain characteristics and trend analysis, and optimizing resource configuration and performance optimization decisions. By classifying private management domains, customized resource allocation solutions can be provided for different private management domains based on differences in regions and business types, thereby better meeting the needs of different management domains and improving cloud resource utilization efficiency and user satisfaction.

[0026] In another embodiment, based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the new management domain is identified and classified using the management domain group, so as to select a management domain group matching the new management domain; and based on the region and business type of the new management domain and the existing data of the new management domain and the historical data of the matching management domain group, cloud resource demand prediction is performed, including: Analyze the work log of the newly managed management domain to obtain the expected configuration cloud resource type information of the computing tasks currently executed by the newly managed management domain; compare the expected configuration cloud resource type information with the cloud resource type information that all management domain groups can provide for configuration, and select the management domain group that matches the newly managed management domain; then predict the cloud resource demand based on the location and business type of the newly managed management domain and the existing data of the new management domain and the historical data of the matching management domain group.

[0027] The beneficial effect of the above embodiment is that the newly managed management domain will have demands for different types of cloud resources such as computing power cloud resources, cloud memory resources, storage cloud resources and network bandwidth cloud resources during the execution of computing tasks. In order to ensure that the newly managed management domain can obtain appropriate cloud resource configuration at each process stage of the computing task, the work log of the newly managed management domain is analyzed to obtain the expected configuration cloud resource type information of the currently executed computing task; the expected configuration cloud resource type information is compared with the cloud resource type information that all management domain groups can provide for configuration, and a management domain group that matches the newly managed management domain is selected, so that the selected management domain group can configure appropriate cloud resources in the corresponding process stage of the computing task of the newly managed management domain. Cloud resource demand prediction is then performed based on the location and business type of the newly managed management domain and the existing data of the new management domain and the historical data of the matching management domain group. It also monitors the computing tasks currently executed by the newly managed management domain to obtain the task running process information corresponding to the currently executed computing tasks, and then compares the task running process information with the idle cloud resources of all private cloud management domains under the matching management domain cluster to determine the private cloud management domain that can configure the corresponding type of cloud resources for the newly managed management domain, and extracts the corresponding cloud resources from the determined private cloud management domain and configures them for the newly managed management domain, ensuring that the newly managed management domain obtains reliable and continuous cloud resource configuration, improves task operation efficiency and makes full use of the cloud resources of the management domain.

[0028] In another embodiment, the operation quality of the acquisition process of the work log of the newly managed management domain is determined, including: Extract the acquisition time of the work log; Compare the acquisition time of the work log with a preset acquisition time threshold; When the acquisition time of the work log exceeds the preset acquisition time threshold, the operation parameters for acquiring the work log are retrieved, wherein the operation parameters include the diary screening accuracy, the diary retrieval garbled code ratio and the diary retrieval response time; The diary screening accuracy, diary retrieval garbled code ratio and diary retrieval response time are used to obtain a diary retrieval operation evaluation coefficient; The diary retrieval operation evaluation coefficient is obtained by the following formula: Where W represents the diary retrieval operation evaluation coefficient; n represents the number of diary retrieval times; P i represents the diary screening accuracy corresponding to the i-th diary retrieval; L i represents the proportion of garbled characters in the diary retrieval corresponding to the i-th diary retrieval; T i T represents the diary retrieval response time corresponding to the i-th diary retrieval; Lmax Indicates the diary retrieval response time corresponding to the maximum value of the diary retrieval garbled code ratio; T Lmin It indicates the diary retrieval response time corresponding to the minimum value of the diary retrieval garbled code ratio; S indicates the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Among them, S represents the adjustment coefficient; L pmin Indicates the proportion of garbled characters in diary retrieval corresponding to the minimum diary screening accuracy; L pmax Indicates the proportion of garbled characters in diary retrieval corresponding to the maximum diary screening accuracy; L min Indicates the minimum value of the garbled character ratio in diary retrieval; L max Indicates the maximum ratio of garbled characters retrieved from the diary; Comparing the diary retrieval operation evaluation coefficient with a preset coefficient threshold; When the diary retrieval operation evaluation coefficient is lower than a preset coefficient threshold, it is determined that there is an abnormality in the work log acquisition process, and an abnormality alarm is issued.

[0029] The beneficial effect of the above embodiment is that by extracting the acquisition time of the work log and comparing it with the preset acquisition time threshold, real-time monitoring of the work log acquisition process of the newly managed management domain can be achieved. Once the acquisition time exceeds the threshold, further analysis and judgment processes are triggered immediately to ensure that potential anomalies can be discovered and handled in a timely manner. When the acquisition time is abnormal, the scheme not only focuses on a single time indicator, but also further retrieves the operating parameters for obtaining the work log, including the diary screening accuracy, the diary retrieval garbled ratio and the diary retrieval response time. These parameters together constitute a comprehensive evaluation of the operating quality of the work log acquisition process, which helps to identify the problem more accurately. By introducing the diary retrieval operation evaluation coefficient W, the scheme realizes a quantitative evaluation of the operating quality of the work log acquisition process. This coefficient takes into account multiple operating parameters and is calculated by a specific formula, which can more intuitively reflect the operating status of the system. The introduction of the adjustment coefficient S makes the calculation of the evaluation coefficient W more flexible and adaptive. The value of S is adjusted according to the changes in the diary screening accuracy and the diary retrieval garbled ratio, which can more accurately reflect the operating quality of the system in different states. When the log retrieval operation evaluation coefficient is lower than the preset coefficient threshold, the solution can automatically determine that there is an abnormality in the work log acquisition process and issue an abnormality alarm. This helps operation and maintenance personnel to discover problems in a timely manner and take appropriate measures to troubleshoot and repair them, thereby ensuring the stable operation of the system.

[0030] In summary, this technical solution effectively improves the operation quality and reliability of the work log acquisition process of the newly managed domain through real-time monitoring, multi-dimensional evaluation, quantitative evaluation, adaptive adjustment and abnormal alarm.

[0031] See also Figure 2 As shown, a cloud resource configuration system based on multi-private cloud management domain classification is provided in one embodiment of the present application. The cloud resource configuration system based on multi-private cloud management domain classification includes: A historical data acquisition and preprocessing module is used to acquire the cloud resource configuration historical data of all private cloud management domains based on the cloud resource configuration historical records of each of the several private cloud management domains, and to preprocess the cloud resource configuration historical data; A historical data correction module is used to perform data correction processing on the cloud resource configuration historical data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data; The cloud resource configuration feature recognition module is used to perform cluster analysis on the valid cloud resource configuration record data to obtain the cloud resource configuration feature information of all private cloud management domains; The management domain group division module is used to identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions; A management domain group selection module is used to determine the cloud resource demand attribute information of the new management domain based on the work log of the new management domain, and use the management domain group to identify and classify the new management domain, so as to select a management domain group that matches the new management domain; The cloud resource configuration prediction module is used to predict cloud resource demand based on the location and business type of the new management domain and the existing data of the new management domain and the historical data of the matched management domain group.

[0032] The beneficial effects of the above embodiments are as follows: the cloud resource configuration system based on multi-private cloud management domain classification obtains the cloud resource configuration historical data of all private cloud management domains from the cloud resource configuration historical records of several private cloud management domains, and performs preprocessing and data correction on the cloud resource configuration historical data to obtain effective cloud resource configuration record data; then cluster analysis is performed on the effective cloud resource configuration record data, and based on the cloud resource configuration feature information of all private cloud management domains, the similarities and differences between different management domains are accurately identified and analyzed, thereby dividing all private cloud management domains into several management domain groups, which is convenient for the subsequent use of management domain groups to centrally and uniformly provide cloud resource configuration; based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the management domain group is used to identify and classify the newly managed management domain, thereby selecting a management domain group that matches the newly managed management domain, and based on the location and business type of the newly managed management domain, accurate and efficient cloud resource demand prediction is performed according to the existing data of the newly managed management domain and the historical data of the matched management domain group.

[0033] In another embodiment, the historical data acquisition and preprocessing module is used to acquire the cloud resource configuration historical data of all private cloud management domains based on the cloud resource configuration history records of the respective private cloud management domains, and preprocess the cloud resource configuration historical data, including: Obtain the cloud resource configuration history records of several private cloud management domains in a preset historical time interval, analyze the cloud resource configuration history records, and obtain all cloud resource configuration valid behaviors that occurred in each private cloud management domain in the preset historical time interval; analyze all cloud resource configuration valid behaviors to determine the configuration history data of each private cloud management domain on different types of cloud resources; wherein the configuration history data at least includes the configuration occurrence time and the number of configured resources of the corresponding type of cloud resources; the different types of cloud resources include computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources; and perform noise removal preprocessing and missing data completion preprocessing on the configuration history data in turn; The historical data correction module is used to perform data correction processing on the cloud resource configuration historical data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data, including: Based on the Holt-Winters model, data smoothing is performed on the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing. Based on the result of the data smoothing, deviation and abnormal data component elimination and correction processing is performed on the configuration history data to obtain effective cloud resource configuration record data.

[0034] The beneficial effect of the above embodiment is that different private cloud management domains can perform configuration operations of different types of cloud resources, and the different types of cloud resources may include but are not limited to computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources, etc. Different private cloud management domains usually tend to concentrate on configuring one type of cloud resources in actual cloud resource configuration operations. In order to accurately identify the cloud resource configuration characteristics of all private cloud management domains, the cloud resource configuration history records of several private cloud management domains in a preset historical time interval are obtained and analyzed to obtain all valid cloud resource configuration behaviors that occurred in each private cloud management domain in the preset historical time interval, wherein the valid cloud resource configuration behavior may be but is not limited to a behavior in which the cloud resource configuration duration is greater than a preset time length threshold; and the valid cloud resource configuration behavior is analyzed to determine the configuration history data of each private cloud management domain on different types of cloud resources, so that the cloud resource configuration tendency of each private cloud management domain can be quantitatively collected, providing sufficient and comprehensive data support for the subsequent similarity relationship analysis between different private cloud management domains. In addition, the configuration history data is subjected to noise removal preprocessing and missing data completion preprocessing in sequence, so as to remove the noise interference components in the configuration history data and improve the completeness of the data; and based on the Holt-Winters model, the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing is subjected to data smoothing processing; and based on the result of the data smoothing processing, the configuration history data is subjected to deviation and abnormal data component elimination and correction processing, so as to obtain effective cloud resource configuration record data, thereby improving data quality and accuracy, and enhancing the reliability and accuracy of subsequent similarity relationship identification between different private cloud management domains.

[0035] In another embodiment, the cloud resource configuration feature identification module is used to perform cluster analysis on the effective cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains, including: Based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources; The management domain group division module is used to identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions, including: Based on the demand change trend characteristic information, determine the cloud resource type information that is matched and configured for all private cloud management domains; then, based on the cloud resource type information that is matched and configured, divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group are matched and configured with the same type of cloud resources, and the demand change trends of the cloud resources that are matched and configured for all private cloud management domains under each management domain group are within a preset quantity range.

[0036] The beneficial effect of the above embodiment is that each private cloud management domain can simultaneously manage computing power cloud resources, cloud memory resources, storage cloud resources, network bandwidth cloud resources, etc. However, in actual cloud resource configuration operations, each private cloud management domain does not necessarily configure all types of cloud resources managed by its subordinates equally, that is, each private cloud management domain may tend to mainly configure one or two of the cloud resources. At this level, all private cloud management domains can be distinguished by similarity. To this end, based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources. Based on the demand change trend characteristic information, the cloud resource type information of all private cloud management domains is determined, so that all private cloud management domains are divided into several management domain groups; among which, all private cloud management domains under each management domain group match and configure the same type of cloud resources, and the demand change trend of cloud resources matched and configured by all private cloud management domains under each management domain group is within the preset quantity range, so that all private cloud management domains under the same management domain group tend to concentrate on configuring the same type of cloud resources. Through cluster analysis based on the hybrid algorithm of DBSCAN and Kmeans, the similarities and differences between different private management domains can be accurately identified, providing more accurate management domain characteristics and trend analysis, and optimizing resource configuration and performance optimization decisions. By classifying private management domains, customized resource allocation solutions can be provided for different private management domains based on differences in regions and business types, thereby better meeting the needs of different management domains and improving cloud resource utilization efficiency and user satisfaction.

[0037] In another embodiment, the management domain group selection module is used to determine the cloud resource demand attribute information of the new management domain based on the work log of the new management domain, and use the management domain group to identify and classify the new management domain, so as to select the management domain group matching the new management domain, including: Analyze the work log of the newly managed management domain to obtain the expected configuration cloud resource type information of the computing tasks currently executed by the newly managed management domain; compare the expected configuration cloud resource type information with the cloud resource type information that all management domain groups can provide for configuration, and select the management domain group that matches the newly managed management domain.

[0038] The beneficial effect of the above embodiment is that the newly managed management domain will have demands for different types of cloud resources such as computing power cloud resources, cloud memory resources, storage cloud resources and network bandwidth cloud resources during the execution of computing tasks. In order to ensure that the newly managed management domain can obtain appropriate cloud resource configuration at each process stage of the computing task, the work log of the newly managed management domain is analyzed to obtain the expected configuration cloud resource type information of the currently executed computing task; the expected configuration cloud resource type information is compared with the cloud resource type information that all management domain groups can provide for configuration, and a management domain group that matches the newly managed management domain is selected, so that the selected management domain group can configure appropriate cloud resources in the corresponding process stage of the computing task of the newly managed management domain. Cloud resource demand prediction is then performed based on the location and business type of the newly managed management domain and the existing data of the new management domain and the historical data of the matching management domain group. It also monitors the computing tasks currently executed by the newly managed management domain to obtain the task running process information corresponding to the currently executed computing tasks, and then compares the task running process information with the idle cloud resources of all private cloud management domains under the matching management domain cluster to determine the private cloud management domain that can configure the corresponding type of cloud resources for the newly managed management domain, and extracts the corresponding cloud resources from the determined private cloud management domain and configures them for the newly managed management domain, ensuring that the newly managed management domain obtains reliable and continuous cloud resource configuration, improves task operation efficiency and makes full use of the cloud resources of the management domain.

[0039] In another embodiment, the operation quality of the acquisition process of the work log of the newly managed management domain is determined, including: Extract the acquisition time of the work log; Compare the acquisition time of the work log with a preset acquisition time threshold; When the acquisition time of the work log exceeds the preset acquisition time threshold, the operation parameters for acquiring the work log are retrieved, wherein the operation parameters include the diary screening accuracy, the diary retrieval garbled code ratio and the diary retrieval response time; The diary screening accuracy, diary retrieval garbled code ratio and diary retrieval response time are used to obtain a diary retrieval operation evaluation coefficient; The diary retrieval operation evaluation coefficient is obtained by the following formula: Where W represents the diary retrieval operation evaluation coefficient; n represents the number of diary retrieval times; P i represents the diary screening accuracy corresponding to the i-th diary retrieval; L i represents the proportion of garbled characters in the diary retrieval corresponding to the i-th diary retrieval; T i T represents the diary retrieval response time corresponding to the i-th diary retrieval; Lmax Indicates the diary retrieval response time corresponding to the maximum value of the diary retrieval garbled code ratio; T Lmin It indicates the diary retrieval response time corresponding to the minimum value of the diary retrieval garbled code ratio; S indicates the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Among them, S represents the adjustment coefficient; L pmin Indicates the proportion of garbled characters in diary retrieval corresponding to the minimum diary screening accuracy; L pmax Indicates the proportion of garbled characters in diary retrieval corresponding to the maximum diary screening accuracy; L min Indicates the minimum value of the garbled character ratio in diary retrieval; L max Indicates the maximum ratio of garbled characters retrieved from the diary; Comparing the diary retrieval operation evaluation coefficient with a preset coefficient threshold; When the diary retrieval operation evaluation coefficient is lower than a preset coefficient threshold, it is determined that there is an abnormality in the work log acquisition process, and an abnormality alarm is issued.

[0040] The beneficial effect of the above embodiment is that by extracting the acquisition time of the work log and comparing it with the preset acquisition time threshold, real-time monitoring of the work log acquisition process of the newly managed management domain can be achieved. Once the acquisition time exceeds the threshold, further analysis and judgment processes are triggered immediately to ensure that potential anomalies can be discovered and handled in a timely manner. When the acquisition time is abnormal, the scheme not only focuses on a single time indicator, but also further retrieves the operating parameters for obtaining the work log, including the diary screening accuracy, the diary retrieval garbled ratio and the diary retrieval response time. These parameters together constitute a comprehensive evaluation of the operating quality of the work log acquisition process, which helps to identify the problem more accurately. By introducing the diary retrieval operation evaluation coefficient W, the scheme realizes a quantitative evaluation of the operating quality of the work log acquisition process. This coefficient takes into account multiple operating parameters and is calculated by a specific formula, which can more intuitively reflect the operating status of the system. The introduction of the adjustment coefficient S makes the calculation of the evaluation coefficient W more flexible and adaptive. The value of S is adjusted according to the changes in the diary screening accuracy and the diary retrieval garbled ratio, which can more accurately reflect the operating quality of the system in different states. When the log retrieval operation evaluation coefficient is lower than the preset coefficient threshold, the solution can automatically determine that there is an abnormality in the work log acquisition process and issue an abnormality alarm. This helps operation and maintenance personnel to discover problems in a timely manner and take appropriate measures to troubleshoot and repair them, thereby ensuring the stable operation of the system.

[0041] In summary, this technical solution effectively improves the operation quality and reliability of the work log acquisition process of the newly managed domain through real-time monitoring, multi-dimensional evaluation, quantitative evaluation, adaptive adjustment and abnormal alarm.

[0042] In general, the cloud resource configuration method and system based on multi-private cloud management domain classification obtains the cloud resource configuration historical data of all private cloud management domains from the cloud resource configuration historical records of several private cloud management domains, and pre-processes and corrects the cloud resource configuration historical data to obtain effective cloud resource configuration record data; then cluster analysis is performed on the effective cloud resource configuration record data, and based on the cloud resource configuration feature information of all private cloud management domains, the similarities and differences between different management domains are accurately identified and analyzed, so as to divide all private cloud management domains into several management domain groups, so as to facilitate the subsequent use of management domain groups to centrally and uniformly provide cloud resource configuration; based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the management domain group is used to identify and classify the newly managed management domain, so as to select the management domain group that matches the newly managed management domain, and based on the location and business type of the newly managed management domain, accurate and efficient cloud resource demand prediction is performed according to the existing data of the newly managed management domain and the historical data of the matched management domain group.

[0043] The above is only a specific implementation of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the protection scope of the present invention.

Claims

1. A cloud resource configuration method based on multi-private cloud management domain classification, characterized in that: include: Based on the cloud resource configuration history records of the respective private cloud management domains, obtaining the cloud resource configuration history data of all the private cloud management domains, and preprocessing the cloud resource configuration history data; Performing data correction processing on the cloud resource configuration history data that has completed the preprocessing, thereby obtaining valid cloud resource configuration record data; Performing cluster analysis on the effective cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains; based on the cloud resource configuration feature information, identifying and analyzing the similarities and differences between different management domains, and dividing all private cloud management domains into a number of management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions; Based on the work log of the newly managed management domain, determine the cloud resource demand attribute information of the newly managed management domain, use the management domain group to identify and classify the newly managed management domain, so as to select the management domain group that matches the newly managed management domain; then, based on the location and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matching management domain group, perform cloud resource demand prediction.

2. The cloud resource configuration method based on multi-private cloud management domain classification according to claim 1, characterized in that: Based on the cloud resource configuration history records of the respective private cloud management domains, obtaining the cloud resource configuration history data of all the private cloud management domains, and preprocessing the cloud resource configuration history data; Performing data correction processing on the cloud resource configuration history data that has completed the preprocessing, thereby obtaining valid cloud resource configuration record data, including: Obtain cloud resource configuration history records of several private cloud management domains in a preset historical time interval, analyze the cloud resource configuration history records, and obtain all cloud resource configuration valid behaviors that occurred in each private cloud management domain in the preset historical time interval; analyze all cloud resource configuration valid behaviors to determine the configuration history data of each private cloud management domain on different types of cloud resources; wherein the configuration history data at least includes the configuration occurrence time and the number of configured resources of the corresponding type of cloud resources; the different types of cloud resources include computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources; and perform noise removal preprocessing and missing data completion preprocessing on the configuration history data in turn; Based on the Holt-Winters model, data smoothing processing is performed on the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing; then based on the result of the data smoothing processing, deviation and abnormal data component elimination and correction processing is performed on the configuration history data, so as to obtain effective cloud resource configuration record data.

3. The cloud resource configuration method based on multi-private cloud management domain classification according to claim 2, characterized in that: The effective cloud resource configuration record data is clustered and analyzed to obtain cloud resource configuration feature information of all private cloud management domains; based on the cloud resource configuration feature information, similarities and differences between different management domains are identified and analyzed, and all private cloud management domains are divided into several management domain groups; wherein all private cloud management domains under each management domain group meet corresponding cloud resource configuration and change trend similarity conditions, including: Based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources; Based on the demand change trend characteristic information, determine the cloud resource type information matched and configured by all private cloud management domains; then based on the matched and configured cloud resource type information, divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group are matched and configured with the same type of cloud resources, and the demand change trends of the cloud resources matched and configured by all private cloud management domains under each management domain group are within a preset quantity range.

4. The cloud resource configuration method based on multi-private cloud management domain classification according to claim 3, characterized in that: Based on the work log of the newly managed management domain, the cloud resource demand attribute information of the newly managed management domain is determined, and the newly managed management domain is identified and classified by using the management domain group, so as to select the management domain group matching the newly managed management domain; and based on the location and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matching management domain group, cloud resource demand prediction is performed, including: Analyze the work log of the newly managed management domain to obtain the expected configuration cloud resource type information of the computing task currently executed by the newly managed management domain; compare the expected configuration cloud resource type information with the cloud resource type information that all management domain groups can provide for configuration, and select the management domain group that matches the newly managed management domain; then, based on the location and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matching management domain group, predict the cloud resource demand.

5. The cloud resource configuration method based on multi-private cloud management domain classification according to claim 4, characterized in that: The operation quality of the acquisition process of the work log of the newly managed management domain is determined, including: Extract the acquisition time of the work log; Compare the acquisition time of the work log with a preset acquisition time threshold; When the acquisition time of the work log exceeds the preset acquisition time threshold, the operation parameters for acquiring the work log are retrieved, wherein the operation parameters include the diary screening accuracy, the diary retrieval garbled code ratio and the diary retrieval response time; The diary screening accuracy, diary retrieval garbled code ratio and diary retrieval response time are used to obtain a diary retrieval operation evaluation coefficient; The diary retrieval operation evaluation coefficient is obtained by the following formula: Where W represents the diary retrieval operation evaluation coefficient; n represents the number of diary retrieval times; P i represents the diary screening accuracy corresponding to the i-th diary retrieval; L i represents the proportion of garbled characters in the diary retrieval corresponding to the i-th diary retrieval; T i T represents the diary retrieval response time corresponding to the i-th diary retrieval; Lmax Indicates the diary retrieval response time corresponding to the maximum value of the diary retrieval garbled code ratio; T Lmin It indicates the diary retrieval response time corresponding to the minimum value of the diary retrieval garbled code ratio; S indicates the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Among them, S represents the adjustment coefficient; L pmin Indicates the proportion of garbled characters in diary retrieval corresponding to the minimum diary screening accuracy; L pmax Indicates the proportion of garbled characters in diary retrieval corresponding to the maximum diary screening accuracy; L min Indicates the minimum value of the garbled character ratio in diary retrieval; L max Indicates the maximum ratio of garbled characters retrieved from the diary; Comparing the diary retrieval operation evaluation coefficient with a preset coefficient threshold; When the diary retrieval operation evaluation coefficient is lower than a preset coefficient threshold, it is determined that there is an abnormality in the work log acquisition process, and an abnormality alarm is issued.

6. A cloud resource configuration system based on multi-private cloud management domain classification, characterized in that: include: A historical data acquisition and preprocessing module, used to acquire the cloud resource configuration historical data of all private cloud management domains based on the cloud resource configuration historical records of the respective private cloud management domains, and preprocess the cloud resource configuration historical data; A historical data correction module, used to perform data correction processing on the cloud resource configuration historical data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data; A cloud resource configuration feature identification module is used to perform cluster analysis on the valid cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains; A management domain group division module is used to identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions; A management domain group selection module is used to determine the cloud resource demand attribute information of the newly managed management domain based on the work log of the newly managed management domain, identify and classify the newly managed management domain using the management domain group, and select a management domain group that matches the newly managed management domain; The cloud resource configuration prediction module is used to predict cloud resource demand based on the region and business type of the newly managed management domain and the existing data of the newly managed management domain and the historical data of the matched management domain group.

7. The cloud resource configuration system based on multi-private cloud management domain classification according to claim 6, characterized in that: The historical data acquisition and preprocessing module is used to acquire the cloud resource configuration historical data of all private cloud management domains based on the cloud resource configuration history records of the respective private cloud management domains, and preprocess the cloud resource configuration historical data, including: Obtain cloud resource configuration history records of several private cloud management domains in a preset historical time interval, analyze the cloud resource configuration history records, and obtain all cloud resource configuration valid behaviors that occurred in each private cloud management domain in the preset historical time interval; analyze all cloud resource configuration valid behaviors to determine the configuration history data of each private cloud management domain on different types of cloud resources; wherein the configuration history data at least includes the configuration occurrence time and the number of configured resources of the corresponding type of cloud resources; the different types of cloud resources include computing power cloud resources, memory cloud resources, storage cloud resources and network bandwidth cloud resources; and perform noise removal preprocessing and missing data completion preprocessing on the configuration history data in turn; The historical data correction module is used to perform data correction processing on the cloud resource configuration historical data that has completed the preprocessing, so as to obtain valid cloud resource configuration record data, including: Based on the Holt-Winters model, data smoothing processing is performed on the configuration history data that has completed the noise removal preprocessing and the missing data completion preprocessing; then based on the result of the data smoothing processing, deviation and abnormal data component elimination and correction processing is performed on the configuration history data, so as to obtain effective cloud resource configuration record data.

8. The cloud resource configuration system based on multi-private cloud management domain classification according to claim 6, characterized in that: The cloud resource configuration feature identification module is used to perform cluster analysis on the valid cloud resource configuration record data to obtain cloud resource configuration feature information of all private cloud management domains, including: Based on the hybrid algorithm of DBSCAN and Kmeans, cluster analysis is performed on the effective cloud resource configuration record data to obtain the demand change trend characteristic information of all private cloud management domains in the process of configuring different types of cloud resources; The management domain group division module is used to identify and analyze the similarities and differences between different management domains based on the cloud resource configuration feature information, and divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group meet the corresponding cloud resource configuration and change trend similarity conditions, including: Based on the demand change trend characteristic information, determine the cloud resource type information matched and configured by all private cloud management domains; then based on the matched and configured cloud resource type information, divide all private cloud management domains into several management domain groups; wherein all private cloud management domains under each management domain group are matched and configured with the same type of cloud resources, and the demand change trends of the cloud resources matched and configured by all private cloud management domains under each management domain group are within a preset quantity range.

9. The cloud resource configuration system based on multi-private cloud management domain classification according to claim 8, characterized in that: The management domain group selection module is used to determine the cloud resource demand attribute information of the new management domain based on the work log of the new management domain, identify and classify the new management domain by using the management domain group, and select the management domain group matching the new management domain, including: Analyze the work log of the newly managed management domain to obtain the expected configuration cloud resource type information of the computing task currently executed by the newly managed management domain; compare the expected configuration cloud resource type information with the cloud resource type information that all management domain groups can provide for configuration, and select the management domain group that matches the newly managed management domain.

10. The cloud resource configuration method based on multi-private cloud management domain classification according to claim 9, characterized in that: The operation quality of the acquisition process of the work log of the newly managed management domain is determined, including: Extract the acquisition time of the work log; Compare the acquisition time of the work log with a preset acquisition time threshold; When the acquisition time of the work log exceeds the preset acquisition time threshold, the operation parameters for acquiring the work log are retrieved, wherein the operation parameters include the diary screening accuracy, the diary retrieval garbled code ratio and the diary retrieval response time; The diary screening accuracy, diary retrieval garbled code ratio and diary retrieval response time are used to obtain a diary retrieval operation evaluation coefficient; The diary retrieval operation evaluation coefficient is obtained by the following formula: Where W represents the diary retrieval operation evaluation coefficient; n represents the number of diary retrieval times; P i represents the diary screening accuracy corresponding to the i-th diary retrieval; L i represents the proportion of garbled characters in the diary retrieval corresponding to the i-th diary retrieval; T i T represents the diary retrieval response time corresponding to the i-th diary retrieval; Lmax Indicates the diary retrieval response time corresponding to the maximum value of the diary retrieval garbled code ratio; T Lmin It indicates the diary retrieval response time corresponding to the minimum value of the diary retrieval garbled code ratio; S indicates the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Among them, S represents the adjustment coefficient; L pmin Indicates the proportion of garbled characters in diary retrieval corresponding to the minimum diary screening accuracy; L pmax Indicates the proportion of garbled characters in diary retrieval corresponding to the maximum diary screening accuracy; L min Indicates the minimum value of the garbled character ratio in diary retrieval; L max Indicates the maximum ratio of garbled characters retrieved from the diary; Comparing the diary retrieval operation evaluation coefficient with a preset coefficient threshold; When the diary retrieval operation evaluation coefficient is lower than a preset coefficient threshold, it is determined that there is an abnormality in the work log acquisition process, and an abnormality alarm is issued.

Images