Data center visual operation and maintenance management system based on digital twin technology

Abstract

The present invention provides a data center visualization operation and maintenance management system based on digital twin technology, which belongs to the field of digital twin technology, and specifically includes: a sampling and analysis module, a priority division module, a server division module, and a model generation module; wherein the sampling and analysis module is responsible for determining the difficulty of data sampling and analysis; the priority division module is responsible for determining the monitoring priority of the server, and determining the online monitoring frequency of different servers and the real-time monitoring server in the server through the monitoring priority; the server division module is responsible for determining the operating status value of the server and screening the server; the model generation module is responsible for generating differentiated digital twin models of the server through the abnormal conditions of the online monitoring data, thereby improving the efficiency of data sampling and analysis.

Description

Technical Field

[0001] The present invention belongs to the field of digital twin technology, and in particular relates to a data center visual operation and maintenance management system based on digital twin technology. Background Art

[0002] With the acceleration of enterprise informatization, especially for large enterprises, the number of data centers that need operation and maintenance management is increasing. At the same time, the number of servers required for operation and maintenance management in different data centers is also very large. Therefore, how to realize visual operation and maintenance management of data centers has become a technical problem that needs to be solved urgently.

[0003] In order to solve the above technical problems, the invention patent CN202111239851.8 "A data center operation and maintenance method based on digital twin technology" generates a digital twin model based on the actual scenario, which can provide an overview of the entire data center, understand the location of each device and the connection status of all devices, and improve the efficiency of operation and maintenance management. However, at the same time, there are the following technical problems:

[0004] For large data centers, the number of servers that need to be operated and managed is large, and there are certain differences in the data communication protocols of different servers, which makes it difficult to monitor and analyze the operating data of the servers on the digital twin platform. Therefore, if the update frequency of differentiated data sampling and analysis cannot be determined based on the differences in the data communication protocols of different servers, it may not be possible to accurately and reliably implement the operation and maintenance management of different servers.

[0005] In response to the above technical problems, the present invention provides a data center visual operation and maintenance management system based on digital twin technology. Summary of the invention

[0006] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0007] A data center visual operation and maintenance management method based on digital twin technology, characterized by specifically including:

[0008] S1 obtains the number of servers managed by the data center, and determines the difficulty of data sampling and analysis based on the communication protocol type and data volume of online monitoring data of different servers. When the difficulty of data sampling and analysis does not meet the requirements, proceed to the next step;

[0009] S2 obtains the historical load rates of different servers, and determines the monitoring priority of the server in combination with the historical operation data of the server with abnormal historical load rates, and determines the online monitoring frequency of different servers and the real-time monitoring server among the servers through the monitoring priority;

[0010] S3 determines the real-time monitoring server associated with the server through the topological relationship of the server, and uses it as the associated server, determines the amount of data processed by the server within a preset period based on the associated server, and determines the operating status value of the server and filters the server in combination with the energy consumption data of the server within the preset period;

[0011] S4 collects the online monitoring data of the screening server in real time, dynamically adjusts the online monitoring frequency of the server according to the operating status value, and generates a differentiated digital twin model of the server through the abnormal conditions of the online monitoring data.

[0012] The beneficial effects of the present invention are:

[0013] 1. The difficulty of data sampling and analysis is determined by combining the number of servers and the communication protocol type and data volume of the online monitoring data of different servers. This takes into account the differences in the difficulty of data sampling and analysis of different servers due to differences in communication protocol types and data volumes, as well as the increased difficulty of data sampling and analysis due to the large number of servers, thereby achieving an accurate assessment of the difficulty of data sampling and analysis in the data center.

[0014] 2. The online monitoring frequency of different servers and the real-time monitoring servers among the servers are determined by monitoring priority, thereby realizing the screening of servers with drastic load rate changes and more load rate anomalies based on the drastic load rate changes of different servers, the duration of the historical load rate abnormality period, and the uniformity of the distribution, thereby realizing differentiated monitoring of different servers.

[0015] 3. By generating differentiated digital twin models of servers based on abnormal conditions in online monitoring data, differentiated digital twin models can be constructed based on the actual operating status of online monitoring data, thus laying the foundation for operation and maintenance personnel to carry out differentiated operation and maintenance management of different servers and improving the efficiency and pertinence of operation and maintenance management.

[0016] A further technical solution is that the communication protocol type of the online monitoring data of the server is determined according to the model of the server.

[0017] A further technical solution is that the historical load rate is determined based on the historical monitoring data of the server, specifically, the amount of data processed by the server at different times is determined based on the historical monitoring data of the server, and the historical load rate is determined in combination with the rated value of the amount of data processed by the server.

[0018] A further technical solution is to determine the online monitoring frequency of different servers and the determination of the real-time monitoring server among the servers through the monitoring priority, specifically including:

[0019] The online monitoring frequency of the online monitoring data corresponding to the server is determined according to the monitoring priority of the server, and when the monitoring priority of the server is a specific monitoring priority, the server is determined to be a real-time monitoring server.

[0020] A further technical solution is that the energy consumption data includes the energy consumption monitoring amount of the server in a preset time period and the change amount of the energy consumption monitoring amount in different adjacent time periods for determination.

[0021] A further technical solution is to dynamically adjust the online monitoring frequency of the server according to the operating status value, specifically including:

[0022] When the operation status value of the server meets the requirements, determine whether the load rate operation status value, the energy consumption operation status value and the energy consumption anomaly all meet the requirements. If so, monitor the online monitoring data through the online monitoring frequency. If not, determine the corresponding monitoring frequency correction amount through the load rate operation status value, the energy consumption operation status value and the energy consumption anomaly respectively, and determine the corrected monitoring frequency of the server for the online monitoring frequency according to the maximum value of the corresponding monitoring frequency correction amount;

[0023] When the operating status value of the server does not meet the requirements, the energy consumption monitoring frequency compensation amount corresponding to the server is determined according to the operating status value, and the corrected monitoring frequency of the server is determined based on the energy consumption monitoring frequency compensation amount and the online monitoring frequency.

[0024] A further technical solution is to generate a differentiated digital twin model of the server based on the abnormal conditions of the online monitoring data, specifically including:

[0025] Determine the abnormal monitoring data of the server according to the abnormal situation of the online monitoring data, and determine the abnormal amount of the monitoring data of the server according to the types and severity of different abnormal monitoring data of the server and the amount of abnormal monitoring data;

[0026] The abnormality level of the server is determined according to the interval in which the abnormal amount of the monitoring data is located, and the appearance color of the digital twin model corresponding to the server is determined based on the abnormality level.

[0027] In a second aspect, the present invention provides a data center visualized operation and maintenance management system based on digital twin technology, which adopts the above-mentioned data center visualized operation and maintenance management method based on digital twin technology, and is characterized in that it specifically includes:

[0028] Sampling analysis module, priority division module, server division module, model generation module;

[0029] The sampling and analysis module is responsible for obtaining the number of servers under the operation and maintenance management of the data center, and determining the difficulty of data sampling and analysis based on the communication protocol type and data volume of online monitoring data of different servers;

[0030] The priority division module is responsible for obtaining the historical load rates of different servers, and determining the monitoring priority of the server in combination with the historical operation data of the server with abnormal historical load rates, and determining the online monitoring frequency of different servers and the real-time monitoring server among the servers through the monitoring priority;

[0031] The server division module is responsible for determining the real-time monitoring server associated with the server through the topological relationship of the server, and using it as the associated server, determining the amount of data processed by the server within a preset period based on the associated server, and determining the operating status value of the server and filtering the server in combination with the energy consumption data of the server within the preset period;

[0032] The model generation module is responsible for collecting the online monitoring data of the screening server in real time, dynamically adjusting the online monitoring frequency of the server according to the operating status value, and generating a differentiated digital twin model of the server through abnormal conditions of the online monitoring data.

[0033] Other features and advantages will be described in the following description, and partly become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0036] Figure 1 It is a flow chart of a data center visual operation and maintenance management method based on digital twin technology;

[0037] Figure 2 It is a flow chart of the method for determining the difficulty of data sampling analysis;

[0038] Figure 3 It is a framework diagram of a data center visualization operation and maintenance management system based on digital twin technology. DETAILED DESCRIPTION

[0039] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0040] The terms "a", "an", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "comprising" and "having" are used to express an open-ended inclusive meaning and mean that additional elements / components / etc. may be present in addition to the listed elements / components / etc.

[0041] Example 1

[0042] To solve the above problems, according to one aspect of the present invention, Figure 1 As shown, according to one aspect of the present invention, a data center visual operation and maintenance management method based on digital twin technology is provided, which is characterized by specifically including:

[0043] S1 obtains the number of servers managed by the data center, and determines the difficulty of data sampling and analysis based on the communication protocol type and data volume of online monitoring data of different servers. When the difficulty of data sampling and analysis does not meet the requirements, proceed to the next step;

[0044] Specifically, the communication protocol type of the online monitoring data of the server is determined according to the model of the server.

[0045] Specific examples, such as Figure 2 As shown, the method for determining the difficulty of the data sampling analysis in the above step S1 is:

[0046] Obtain the number of the servers, and determine whether the data sampling of the data center is complex based on the number of the servers. If so, proceed to the next step. If not, determine whether the difficulty of data sampling analysis of the data center meets the requirements.

[0047] Determine the data parsing difficulty of the online monitoring data of the server by using the communication protocol type and data volume of the online monitoring data of the server, and determine the complex parsing server of the data center based on the data parsing difficulty, and judge whether the data sampling of the data center is complex by using the complex parsing server and the number of servers. If so, proceed to the next step; if not, determine whether the data sampling analysis difficulty of the data center meets the requirements;

[0048] Obtain the number of servers in the data center and the data parsing difficulty of different servers, and determine the comprehensive data parsing difficulty of the data center in combination with the number of communication protocol types of the servers in the data center and the amount of online monitoring data, and judge whether the comprehensive data parsing difficulty meets the requirements. If so, determine that the data sampling and analysis difficulty of the data center meets the requirements. If not, proceed to the next step;

[0049] The number of complex analysis servers and the data analysis difficulty of the data center are obtained, and the data sampling and analysis difficulty of the data center is determined based on the amount of online monitoring data of the complex analysis servers and the comprehensive data analysis difficulty of the data center.

[0050] In another possible embodiment, the method for determining the difficulty of the data sampling analysis in the above step S1 is:

[0051] Obtain the number of servers in the data center and the data parsing difficulty of different servers, and determine the comprehensive data parsing difficulty of the data center in combination with the number of communication protocol types of the servers in the data center and the amount of online monitoring data, and judge whether the comprehensive data parsing difficulty meets the requirements. If so, determine that the data sampling and analysis difficulty of the data center meets the requirements. If not, proceed to the next step;

[0052] Determine the data parsing difficulty of the online monitoring data of the server by using the communication protocol type and data volume of the online monitoring data of the server, and determine the complex parsing server of the data center based on the data parsing difficulty;

[0053] The number of complex analysis servers and the data analysis difficulty of the data center are obtained, and the data sampling and analysis difficulty of the data center is determined based on the amount of online monitoring data of the complex analysis servers and the comprehensive data analysis difficulty of the data center.

[0054] It should be noted that when the difficulty of data sampling and analysis meets the requirements, all servers are directly monitored in real time, and differentiated digital twin models of the servers are generated based on the abnormal conditions of the online monitoring data.

[0055] S2 obtains the historical load rates of different servers, and determines the monitoring priority of the server in combination with the historical operation data of the server with abnormal historical load rates, and determines the online monitoring frequency of different servers and the real-time monitoring server among the servers through the monitoring priority;

[0056] It can be understood that the historical load rate is determined based on the historical monitoring data of the server. Specifically, the amount of data processed by the server at different times is determined based on the historical monitoring data of the server, and the historical load rate is determined in combination with the rated value of the amount of data processed by the server.

[0057] Specifically, the method for determining the monitoring priority of the server in step S2 is as follows:

[0058] Determine the operating time when the historical load rate of the server is abnormal by using the historical load rate of the server and the preset load rate threshold, and determine whether the online monitoring data of the server is stable according to the proportion of the operating time when the historical load rate of the server is abnormal. If so, proceed to the next step; if not, determine that the monitoring priority of the server is the first level;

[0059] Determine the overload operation time of the server through the operation time when the historical load rate of the server is abnormal, and determine whether the online monitoring data of the server is stable based on the proportion of the overload operation time of the server, if so, proceed to the next step, if not, determine that the monitoring priority of the server is the first level;

[0060] The operation time when the historical load rate of the server is abnormal is taken as the abnormal operation time, and the abnormal operation time period is determined according to the number of abnormal operation times of the server in different unit time lengths within the preset time, and whether the online monitoring data of the server is stable is determined by the number of abnormal operation time periods of the server within the preset time, if so, proceed to the next step, if not, determine that the monitoring priority of the server is the first level;

[0061] Obtain the number and proportion of abnormal operation time periods of the server within a preset time, and determine the load rate fluctuation amount of the server in combination with the load rate of the server in different abnormal operation time periods and the number of abnormal operation moments;

[0062] The overloaded operation time and the proportion of the overloaded operation time of the server are obtained, and the monitoring priority of the server is determined based on the abnormal operation time of the historical load rate, the proportion of the operation time and the load rate fluctuation of the server.

[0063] Specifically, the monitoring priority of the server includes at least three levels, wherein the monitoring priority of the first level is greater than that of the second level, and the monitoring priority of the second level is greater than that of the third level.

[0064] In another embodiment, the method for determining the monitoring priority of the server in step S2 is:

[0065] Determine the operating time when the historical load rate of the server is abnormal by using the historical load rate of the server and the preset load rate threshold, take the operating time when the historical load rate of the server is abnormal as the abnormal operating time, and determine the abnormal operating period by the number of abnormal operating times in different unit time lengths within the preset time according to the abnormal operating time;

[0066] When the proportion of abnormal operation time periods of the server within the preset time does not meet the requirement, determining the monitoring priority of the server to be the first level;

[0067] When the proportion of abnormal operation time periods of the server within the preset time meets the requirement, the overload operation time of the server is determined by the abnormal operation time of the server, and the overload operation time period is determined by the number of overload operation moments existing in different unit time lengths within the preset time according to the overload operation time;

[0068] When the proportion of the overload operation period of the server within the preset time does not meet the requirement, determining the monitoring priority of the server to be the first level;

[0069] When the proportion of the overloaded operation time periods of the server within the preset time meets the requirement, the number and proportion of the abnormal operation time periods of the server within the preset time are obtained, and the load rate fluctuation amount of the server is determined in combination with the load rate of the server in different abnormal operation time periods and the number of abnormal operation moments;

[0070] Obtain the number and proportion of overloaded operation time periods of the server within a preset time, and determine the load rate operation reliability of the server in combination with the load rate of the server in different overloaded operation time periods and the number of overloaded operation moments;

[0071] Obtain the overloaded operating time and the proportion of the overloaded operating time of the server, and determine the basic monitoring priority based on the historical abnormal operating time and the proportion of the operating time; determine the monitoring priority of the server based on the load rate fluctuation amount, load rate, operating reliability and the basic monitoring priority of the server.

[0072] Specifically, determining the online monitoring frequency of different servers and determining the real-time monitoring server among the servers according to the monitoring priority specifically includes:

[0073] The online monitoring frequency of the online monitoring data corresponding to the server is determined according to the monitoring priority of the server, and when the monitoring priority of the server is a specific monitoring priority, the server is determined to be a real-time monitoring server.

[0074] S3 determines the real-time monitoring server associated with the server through the topological relationship of the server, and uses it as the associated server, determines the amount of data processed by the server within a preset period based on the associated server, and determines the operating status value of the server and filters the server in combination with the energy consumption data of the server within the preset period;

[0075] Furthermore, the energy consumption data includes the energy consumption monitoring amount of the server in a preset time period and the change amount of the energy consumption monitoring amount in different adjacent time periods.

[0076] Specifically, the method for determining the operating status value in step S3 is as follows:

[0077] Determining the load rate of the server at different times in the preset period of time by the amount of data processed by the server in the preset period of time, and determining whether the operation state of the server is stable based on the load rate of the server at different times in the preset period of time, if yes, proceeding to the next step, if not, determining that the server is a screening server;

[0078] Determine the energy consumption monitoring amount of the server at different times in the preset period through the energy consumption data of the server in the preset period, and determine whether the operation state of the server is stable based on the energy consumption monitoring amount of the server at different times in the preset period, if yes, proceed to the next step, if not, determine that the server is a screening server;

[0079] Determine the number and duration of the abnormal load rate of the server within the preset period based on the load rate of the server at different times within the preset period, and determine the load rate operating state value of the server in combination with the load rate variation of the server at different adjacent times within the preset period;

[0080] The number and duration of the abnormal energy consumption monitoring amount of the server within the preset period are determined based on the energy consumption detection amount of the server at different times within the preset period, and the energy consumption operation state value of the server is determined in combination with the change amount of the energy consumption monitoring amount of the server at different adjacent times within the preset period;

[0081] The estimated energy consumption at different times is determined by the load rate of the server at different times within a preset time period, and the abnormal energy consumption of the server is determined based on the deviation between the energy consumption monitoring amount at different times within the preset time period and the estimated energy consumption, and the operating status value is determined by the abnormal energy consumption, the load rate operating status value and the energy consumption operating status value.

[0082] In another embodiment, the method for determining the operating status value in step S3 is:

[0083] Determine the number and duration of abnormal load rate moments of the server within the preset period based on the load rate of the server at different times within the preset period, determine the load rate operation status value of the server in combination with the load rate change at different adjacent times within the preset period, and determine whether the operation status of the server is stable based on the load rate operation status value of the server within the preset period, if yes, proceed to the next step, if not, determine that the server is a screening server;

[0084] Based on the energy consumption detection amount of the server at different moments in the preset time period, the number and duration of the moments when the energy consumption monitoring amount of the server in the preset time period is abnormal are determined, and in combination with the change amount of the energy consumption monitoring amount of the server at different adjacent moments in the preset time period, the energy consumption operation state value of the server is determined, and based on the energy consumption operation state value of the server in the preset time period, whether the operation state of the server is stable is determined, if yes, proceeding to the next step, if not, determining that the server is a screening server;

[0085] Determine the estimated energy consumption at different times by the load rate of the server at different times within a preset period, determine the abnormal energy consumption of the server according to the deviation between the energy consumption monitoring amount of the server at different times within the preset period and the estimated energy consumption, and determine whether the operation state of the server is reliable based on the abnormal energy consumption of the server, if yes, proceed to the next step, if not, determine that the server is a screening server;

[0086] The operation status value is determined by the energy consumption abnormality, the load rate operation status value and the energy consumption operation status value.

[0087] It is understandable that when the running status value is small, the running status is not stable enough and it is used as a screening server.

[0088] S4 collects the online monitoring data of the screening server in real time, dynamically adjusts the online monitoring frequency of the server according to the operating status value, and generates a differentiated digital twin model of the server through the abnormal conditions of the online monitoring data.

[0089] It should be noted that dynamically adjusting the online monitoring frequency of the server according to the running status value specifically includes:

[0090] When the operation status value of the server meets the requirements, determine whether the load rate operation status value, the energy consumption operation status value and the energy consumption anomaly all meet the requirements. If so, monitor the online monitoring data through the online monitoring frequency. If not, determine the corresponding monitoring frequency correction amount through the load rate operation status value, the energy consumption operation status value and the energy consumption anomaly respectively, and determine the corrected monitoring frequency of the server for the online monitoring frequency according to the maximum value of the corresponding monitoring frequency correction amount;

[0091] When the operating status value of the server does not meet the requirements, the energy consumption monitoring frequency compensation amount corresponding to the server is determined according to the operating status value, and the corrected monitoring frequency of the server is determined based on the energy consumption monitoring frequency compensation amount and the online monitoring frequency.

[0092] Specifically, generating a differentiated digital twin model of the server based on the abnormal situation of the online monitoring data specifically includes:

[0093] Determine the abnormal monitoring data of the server according to the abnormal situation of the online monitoring data, and determine the abnormal amount of the monitoring data of the server according to the types and severity of different abnormal monitoring data of the server and the amount of abnormal monitoring data;

[0094] The abnormality level of the server is determined according to the interval in which the abnormal amount of the monitoring data is located, and the appearance color of the digital twin model corresponding to the server is determined based on the abnormality level.

[0095] Example 2

[0096] On the other hand, Figure 3 As shown, the present invention provides a data center visualization operation and maintenance management system based on digital twin technology, which adopts the above-mentioned data center visualization operation and maintenance management method based on digital twin technology, and is characterized in that it specifically includes:

[0097] Sampling analysis module, priority division module, server division module, model generation module;

[0098] The sampling and analysis module is responsible for obtaining the number of servers under the operation and maintenance management of the data center, and determining the difficulty of data sampling and analysis based on the communication protocol type and data volume of online monitoring data of different servers;

[0099] The priority division module is responsible for obtaining the historical load rates of different servers, and determining the monitoring priority of the server in combination with the historical operation data of the server with abnormal historical load rates, and determining the online monitoring frequency of different servers and the real-time monitoring server among the servers through the monitoring priority;

[0100] The server division module is responsible for determining the real-time monitoring server associated with the server through the topological relationship of the server, and using it as the associated server, determining the amount of data processed by the server within a preset period based on the associated server, and determining the operating status value of the server and filtering the server in combination with the energy consumption data of the server within the preset period;

[0101] The model generation module is responsible for collecting the online monitoring data of the screening server in real time, dynamically adjusting the online monitoring frequency of the server according to the operating status value, and generating a differentiated digital twin model of the server through abnormal conditions of the online monitoring data.

[0102] Through the above embodiments, the applicant has achieved the following technical effects:

[0103] 1. The difficulty of data sampling and analysis is determined by combining the number of servers and the communication protocol type and data volume of the online monitoring data of different servers. This takes into account the differences in the difficulty of data sampling and analysis of different servers due to differences in communication protocol types and data volumes, as well as the increased difficulty of data sampling and analysis due to the large number of servers, thereby achieving an accurate assessment of the difficulty of data sampling and analysis in the data center.

[0104] 2. The online monitoring frequency of different servers and the real-time monitoring servers among the servers are determined by monitoring priority, thereby realizing the screening of servers with drastic load rate changes and more load rate anomalies based on the drastic load rate changes of different servers, the duration of the historical load rate abnormality period, and the uniformity of the distribution, thereby realizing differentiated monitoring of different servers.

[0105] 3. By generating differentiated digital twin models of servers based on abnormal conditions in online monitoring data, differentiated digital twin models can be constructed based on the actual operating status of online monitoring data, thus laying the foundation for operation and maintenance personnel to carry out differentiated operation and maintenance management of different servers and improving the efficiency and pertinence of operation and maintenance management.

[0106] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0107] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.

Claims

1. A data center visual operation and maintenance management method based on digital twin technology, characterized in that: Specifically include: Obtain the number of servers managed by the data center, and determine the difficulty of data sampling and analysis based on the communication protocol type and data volume of online monitoring data of different servers. If the difficulty of data sampling and analysis does not meet the requirements, proceed to the next step. Obtaining historical load rates of different servers, and determining the monitoring priority of the server in combination with historical operation data showing abnormal historical load rates of the server, and determining the online monitoring frequency of different servers and the real-time monitoring server among the servers through the monitoring priority; Determine the real-time monitoring server associated with the server through the topological relationship of the server, and use it as the associated server, determine the amount of data processed by the server within a preset time period based on the associated server, and determine the operating status value of the server and filter the server in combination with the energy consumption data of the server within the preset time period; Collecting online monitoring data of the screening server in real time, dynamically adjusting the online monitoring frequency of the server according to the operating status value, and generating a differentiated digital twin model of the server based on abnormal conditions of the online monitoring data; Dynamically adjusting the online monitoring frequency of the server according to the running status value specifically includes: When the operation status value of the server meets the requirements, determine whether the load rate operation status value, the energy consumption operation status value and the energy consumption anomaly all meet the requirements. If so, monitor the online monitoring data through the online monitoring frequency. If not, determine the corresponding monitoring frequency correction amount through the load rate operation status value, the energy consumption operation status value and the energy consumption anomaly respectively, and determine the corrected monitoring frequency of the server for the online monitoring frequency according to the maximum value of the corresponding monitoring frequency correction amount; When the operating status value of the server does not meet the requirement, determining the energy consumption monitoring frequency compensation amount corresponding to the server according to the operating status value, and determining the corrected monitoring frequency of the server based on the energy consumption monitoring frequency compensation amount and the online monitoring frequency; Generating a differentiated digital twin model of the server based on the abnormal situation of the online monitoring data specifically includes: Determine the abnormal monitoring data of the server according to the abnormal situation of the online monitoring data, and determine the abnormal amount of the monitoring data of the server according to the types and severity of different abnormal monitoring data of the server and the amount of abnormal monitoring data; Determine the abnormality level of the server according to the interval in which the abnormal amount of the monitoring data is located, and determine the appearance color of the digital twin model corresponding to the server based on the abnormality level; The method for determining the difficulty of data sampling analysis is: Obtain the number of servers in the data center and the data parsing difficulty of different servers, and determine the comprehensive data parsing difficulty of the data center in combination with the number of communication protocol types of the servers in the data center and the amount of online monitoring data, and judge whether the comprehensive data parsing difficulty meets the requirements. If so, determine that the data sampling and analysis difficulty of the data center meets the requirements. If not, proceed to the next step; Determine the data parsing difficulty of the online monitoring data of the server by using the communication protocol type and data volume of the online monitoring data of the server, and determine the complex parsing server of the data center based on the data parsing difficulty; The number of complex analysis servers and the data analysis difficulty of the data center are obtained, and the data sampling and analysis difficulty of the data center is determined based on the amount of online monitoring data of the complex analysis servers and the comprehensive data analysis difficulty of the data center.

2. The data center visual operation and maintenance management method according to claim 1, characterized in that: The communication protocol type of the online monitoring data of the server is determined according to the model of the server.

3. The data center visual operation and maintenance management method according to claim 1, characterized in that: The historical load rate is determined based on the historical monitoring data of the server. Specifically, the amount of data processed by the server at different times is determined based on the historical monitoring data of the server, and the historical load rate is determined in combination with the rated value of the amount of data processed by the server.

4. The data center visual operation and maintenance management method according to claim 1, characterized in that: The method for determining the monitoring priority of the server is: Determine the operating time when the historical load rate of the server is abnormal by using the historical load rate of the server and the preset load rate threshold, and determine whether the online monitoring data of the server is stable according to the proportion of the operating time when the historical load rate of the server is abnormal. If so, proceed to the next step; if not, determine that the monitoring priority of the server is the first level; Determine the overload operation time of the server through the operation time when the historical load rate of the server is abnormal, and determine whether the online monitoring data of the server is stable based on the proportion of the overload operation time of the server, if so, proceed to the next step, if not, determine that the monitoring priority of the server is the first level; The operation time when the historical load rate of the server is abnormal is taken as the abnormal operation time, and the abnormal operation time period is determined according to the number of abnormal operation times of the server in different unit time lengths within the preset time, and whether the online monitoring data of the server is stable is determined by the number of abnormal operation time periods of the server within the preset time, if so, proceed to the next step, if not, determine that the monitoring priority of the server is the first level; Obtain the number and proportion of abnormal operation time periods of the server within a preset time, and determine the load rate fluctuation amount of the server in combination with the load rate of the server in different abnormal operation time periods and the number of abnormal operation moments; The overloaded operation time and the proportion of the overloaded operation time of the server are obtained, and the monitoring priority of the server is determined based on the abnormal operation time of the historical load rate, the proportion of the operation time and the load rate fluctuation of the server.

5. The data center visual operation and maintenance management method according to claim 1, characterized in that: The monitoring priority of the server includes at least three levels, wherein the monitoring priority of the first level is greater than the monitoring priority of the second level, and the monitoring priority of the second level is greater than the third level.

6. The data center visual operation and maintenance management method according to claim 1, characterized in that: Determining the online monitoring frequency of different servers and determining the real-time monitoring server among the servers through the monitoring priority specifically includes: The online monitoring frequency of the online monitoring data corresponding to the server is determined according to the monitoring priority of the server, and when the monitoring priority of the server is a specific monitoring priority, the server is determined to be a real-time monitoring server.

7. The data center visual operation and maintenance management method according to claim 1, characterized in that: The energy consumption data includes the energy consumption monitoring amount of the server in a preset time period and the change amount of the energy consumption monitoring amount in different adjacent time periods.

8. The data center visual operation and maintenance management method according to claim 1, characterized in that: The method for determining the operating status value is: Determining the load rate of the server at different times in the preset period of time by the amount of data processed by the server in the preset period of time, and determining whether the operation state of the server is stable based on the load rate of the server at different times in the preset period of time, if yes, proceeding to the next step, if not, determining that the server is a screening server; Determine the energy consumption monitoring amount of the server at different times in the preset period through the energy consumption data of the server in the preset period, and determine whether the operation state of the server is stable based on the energy consumption monitoring amount of the server at different times in the preset period, if yes, proceed to the next step, if not, determine that the server is a screening server; Determine the number and duration of the abnormal load rate of the server within the preset period based on the load rate of the server at different times within the preset period, and determine the load rate operating state value of the server in combination with the load rate variation of the server at different adjacent times within the preset period; The number and duration of the abnormal energy consumption monitoring amount of the server within the preset period are determined based on the energy consumption detection amount of the server at different times within the preset period, and the energy consumption operation state value of the server is determined in combination with the change amount of the energy consumption monitoring amount of the server at different adjacent times within the preset period; The estimated energy consumption at different times is determined by the load rate of the server at different times within a preset time period, and the abnormal energy consumption of the server is determined based on the deviation between the energy consumption monitoring amount at different times within the preset time period and the estimated energy consumption, and the operating status value is determined by the abnormal energy consumption, the load rate operating status value and the energy consumption operating status value.

9. A data center visual operation and maintenance management system based on digital twin technology, adopting a data center visual operation and maintenance management method based on digital twin technology according to any one of claims 1 to 8, characterized in that: Specifically include: Sampling analysis module, priority division module, server division module, model generation module; The sampling and analysis module is responsible for obtaining the number of servers under the operation and maintenance management of the data center, and determining the difficulty of data sampling and analysis based on the communication protocol type and data volume of online monitoring data of different servers; The priority division module is responsible for obtaining the historical load rates of different servers, and determining the monitoring priority of the server in combination with the historical operation data of the server with abnormal historical load rates, and determining the online monitoring frequency of different servers and the real-time monitoring server among the servers through the monitoring priority; The server division module is responsible for determining the real-time monitoring server associated with the server through the topological relationship of the server, and using it as the associated server, determining the amount of data processed by the server within a preset period based on the associated server, and determining the operating status value of the server and filtering the server in combination with the energy consumption data of the server within the preset period; The model generation module is responsible for collecting the online monitoring data of the screening server in real time, dynamically adjusting the online monitoring frequency of the server according to the operating status value, and generating a differentiated digital twin model of the server through abnormal conditions of the online monitoring data.

Images