Management method based on network management system and electronic equipment

Abstract

The invention provides a management method based on a network management system and electronic equipment, and relates to the technical field of communication, the method comprises the steps that a first log set and a second log set are acquired from the network management system, the first log set comprises a plurality of system operation logs, and the second log set comprises a plurality of system operation logs; the second log set comprises at least one resource monitoring log, based on the first log set and the second log set, determining a mapping relationship between an operator and a system resource, the operator being used for realizing an operation executed by the network management system, and based on the mapping relationship between the operator and the system resource, determining the operation executed by the network management system. And scheduling the system resources of the network management system. According to the technical scheme provided by the invention, automatic analysis and optimization of system resources can be realized, and the management capability and user experience of a network management system are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a management method and electronic equipment based on a network management system. Background Art

[0002] The operation support system (OSS) is a network management system used in the telecommunications field and is a necessary support platform for the development and operation of telecommunications services. OSS can be responsible for the inspection and management of the communication quality and operation of the telecommunications network, record and collect various data in the operation of the entire network, and perform monitoring and control functions on various devices (such as base stations, etc.). However, the continuous increase in the scale and new features of telecommunications networks has also brought new challenges to the system capabilities of OSS.

[0003] In the prior art, the system information of each system in the network management system can be obtained, and the user can analyze the system information based on management experience, evaluate the resource consumption of the network management system, and adjust the parameters of each system so that each system can also operate according to the new parameters. However, this management method requires users to learn and accumulate management experience for a long time, is highly subjective, and has poor management accuracy. Especially as the scale of the target network continues to expand, it is impossible to achieve comprehensive management of the network management system based on manual experience alone. Summary of the invention

[0004] In view of this, the present application provides a management method and electronic device based on a network management system, which can realize automated analysis and optimization of system resources in network management systems such as OSS, and significantly improve the management capabilities and user experience of the network management system.

[0005] In order to achieve the above-mentioned objectives, in the first aspect, an embodiment of the present application provides a management method based on a network management system, including: obtaining a first log set and a second log set from the network management system, the first log set including multiple system operation logs, and the second log set including at least one resource monitoring log; based on the first log set and the second log set, determining a mapping relationship between an operator and system resources, the operator being used to implement the operations performed by the network management system; based on the mapping relationship between the operator and the system resources, scheduling the system resources of the network management system.

[0006] In an embodiment of the present application, a first log set and a second log set in a network management system can be obtained, the first log set includes multiple system operation logs, and the second log set includes at least one resource monitoring log. Among them, the system operation log can record the operations performed by the network management system, and the resource monitoring log can record the occupancy of the system resources of the network management system. Therefore, based on the system operation log and the resource monitoring log, a mapping relationship between operators that perform various operations and system resources can be constructed. Based on this mapping relationship, the demand for system resources when the network management system performs various operations can be determined, and therefore, based on the mapping relationship between the operator and the system resources, the system resources of the network management system can be scheduled, thereby realizing automatic analysis and optimization of system resources, significantly improving the management capabilities and user experience of the network management system.

[0007] The network management system can implement a set of functions configured by a user or developer, and the set of functions can include one or more functions. The network management system can be a multi-layer structure, and each layer of the structure can implement at least part of the functions in the set of functions. Exemplarily, in order from high level to low level, the network management system can include system level, subsystem level, module level, service level and operator level, wherein each level can include one or more higher levels. The operator level can include multiple operators, and the operator can serve as a functional unit in the network management system. In some embodiments, the function that an operator can implement can include only one operation. That is, the operator can serve as the smallest functional unit in the network management system, and then as the smallest measurement unit of occupied resources.

[0008] In some implementations, the network management system may include an OSS.

[0009] System operation logs, also called operation logs, can record the operations performed by the network management system.

[0010] Resource monitoring logs can be used to record the usage of system resources.

[0011] In some implementations, in order to improve the accuracy of obtaining system operation logs and resource monitoring logs, at least one of a first target time period, a target system range, and a target resource type may be determined, and a first log set and a second log set may be obtained from a network management system based on at least one of the first target time period, the target system range, and the target resource type. In some implementations, at least one of the first target time period, the target system range, and the target resource type submitted by a user may be received. Of course, in actual applications, at least one of the first target time period, the target system range, and the target resource type may also be determined by other means.

[0012] The first target time period can be used to indicate the time period for obtaining system operation logs and resource monitoring logs. Based on the first target time period, the system operation logs and resource monitoring logs belonging to the first target time period can be obtained. For example, the first target time period may include all time periods before the current moment, within the last 1 day, within a certain day specified in the history, and the like.

[0013] The target system scope can be used to indicate at least one system included in the network management system, that is, to indicate the source of the system operation log and resource monitoring log. Based on the target system scope, the system operation log and resource monitoring log derived from the target system scope can be obtained. For example, the target system scope can include at least one of a device access system, a data processing system, and a forwarding and presentation system.

[0014] The target resource type may be used to indicate the type of system resource recorded in the acquired resource monitoring log.

[0015] In some embodiments, determining the mapping relationship between operators and system resources based on the first log set and the second log set includes: classifying the multiple system operation logs included in the first log set to obtain a first operator set, the first operator set including at least one operator, wherein each of the operators is used to implement a type of operation performed by the network management system, and each of the operators corresponds to at least part of the system operation logs in the first log set; determining the resource occupancy type corresponding to each of the operators in the first operator set based on the second log set, the resource occupancy type being a valid operator, a periodic operator or an invalid operator, wherein the valid operator corresponds to a data analysis type operation, the periodic operator corresponds to a scheduled task type operation, and the invalid operator corresponds to a printing log type operation; determining the mapping relationship between the operator and system resources based on the number of operators of each resource occupancy type in the second log set and the first operator set.

[0016] By classifying the multiple system operation logs included in the first log set, operators for implementing various operations performed by the network management system are identified, and based on the resource monitoring logs included in the second log set, the resource occupancy type corresponding to each operator in the first operator set is further identified, and the resource occupancy type is a valid operator, a periodic operator, or an invalid operator. Based on the number of operators of each resource occupancy type in the second log set and the first operator set, the mapping relationship between the operator and the system resource is determined, thereby realizing the identification and positioning of system resource consumption at the operator level, improving the accuracy of dynamic evaluation and prediction of system resource consumption, and thus improving the accuracy of resource scheduling.

[0017] In some implementations, the classifying the multiple system operation logs included in the first log set to obtain the first operator set includes: classifying the multiple system operation logs based on the string lengths and / or similarities of the multiple system operation logs included in the first log set to obtain the first operator set, wherein the difference between the string lengths of the multiple system operation logs corresponding to the same operator is less than a first value, and the similarity of the multiple system operation logs corresponding to the same operator is greater than or equal to a second value. That is, the lengths of the multiple system operation logs corresponding to the same operator are the same or similar, and the strings included in the multiple system operation logs are the same or similar.

[0018] Exemplarily, the multiple system operation logs in the first set may be partitioned based on the string lengths of the multiple system operation logs to obtain at least one partition, each partition including at least part of the system operation logs in the first set, and the string lengths of the system operation logs included in each partition are the same or similar, and then the system operation logs in each partition may be classified based on the similarity of the system operation logs in each partition to obtain at least one operator, each operator corresponding to at least part of the system operation logs in the first set. Among them, since the possibility of system operation logs with large length differences being recorded as the same type of operation (i.e. corresponding to the same operator) is small, the system operation logs in the first set may be preliminarily classified based on the lengths of the system operation logs, which can significantly improve the efficiency of classifying the system operation logs in the first set.

[0019] In some embodiments, partitioning the multiple system operation logs based on the string lengths of the multiple system operation logs in the first set may include: creating at least one partition, each partition corresponding to a string length, and determining the string lengths of each system operation log in the first set. If the difference between the string length of the system operation log and the string length corresponding to any partition is less than a first value, then determining that the system operation log belongs to any partition. If the difference between the string length of the system operation log and the string length corresponding to each partition is greater than or equal to the first value, then creating a new partition, and determining that the new system operation log belongs to the new partition, and the new partition corresponds to the string length of the system log.

[0020] In some implementations, the similarity between any two system operation logs may be determined based on the longest common subsequence (LCS).

[0021] In some embodiments, before classifying the plurality of system operation logs included in the first log set to obtain the first operator set, the method further includes: deleting preset characters included in each of the system operation logs. Preprocessing the system operation logs may also be referred to as cleaning the system operation logs. Preprocessing can reduce system operation logs that cannot be applied to subsequent steps (i.e., invalid system operation logs), and can improve the accuracy of subsequent operator identification and classification.

[0022] In some implementations, the method further includes: obtaining system operation logs with a number of occurrences greater than or equal to a third value from at least part of the system operation logs corresponding to the operator; and labeling the operator based on the system operation logs with a number of occurrences greater than or equal to the third value. That is, the most or more system operation logs corresponding to the operator are used as labels for the operator, so as to facilitate determining the specific operation and other features corresponding to the operator, and then distinguish and identify different operators.

[0023] In some embodiments, determining the resource occupancy type corresponding to each operator in the first operator set based on the second log set includes: determining the fluctuation characteristics of the occupancy of the system resources in multiple time periods based on the second log set; determining the distribution characteristics of the operators in the multiple time periods; determining the resource occupancy type corresponding to the operator based on the fluctuation characteristics of the occupancy of the system resources in multiple time periods and the distribution characteristics of the operators in the multiple time periods.

[0024] In some embodiments, the total time period of the first log set and the second log set can be obtained, and the total time period can also be called a monitoring period. The total time period is divided into multiple time periods, and the duration of each time period can be the first duration. Based on the first log set and the multiple time periods, the operators included in the network management system in each time period can be determined. By comparing the operators included in the network management system in the multiple time periods, the distribution characteristics of the operators of the network management system in the multiple time periods can be determined, such as whether each operator appears and the frequency of occurrence. Based on the second log set and the multiple time periods, the occupation of system resources of the network management system in each time period can be determined. By comparing the occupation of system resources of the network management system in the multiple time periods, the distribution characteristics of the system resources of the network management system in the multiple time periods can be determined, such as the change of the occupation of system resources. Based on the degree of synchronization between the time period from the appearance to the disappearance of the operator and the time period from the increase to the decrease of the system resources, the resource occupation type of the operator is determined. Exemplarily, during the period of system resource change, the operators that also appear densely are valid operators, the operators whose number remains basically unchanged are periodic operators, and the operators that appear randomly are invalid operators.

[0025] In some embodiments, the method also includes: determining an entropy value corresponding to the operator, the entropy value being used to indicate the randomness of the distribution of the operator in the multiple time periods; and verifying the resource occupancy type corresponding to the operator based on the entropy value corresponding to the operator, thereby cross-validating the classification result based on the entropy value with the classification result based on the resource monitoring log, thereby improving the accuracy of identifying and classifying the operator.

[0026] In some embodiments, when the entropy value of the operator is less than the fourth value (for example, the entropy value is a smaller value such as 0), the resource occupation type of the operator is determined to be a periodic operator; when the entropy value of the operator is greater than or equal to the fourth value and less than the fifth value (that is, the entropy value is a medium value), the resource occupation type of the operator is determined to be a periodic operator; when the entropy value of the operator is greater than or equal to the fifth value (that is, the entropy value is a larger or maximum value), the resource occupation type of the operator is determined to be an invalid operator. Among them, the entropy value corresponding to the operator is used to indicate the randomness of the distribution of the operator in multiple time periods. When the entropy value is larger, it means that the randomness of the operator is smaller; when the entropy value is smaller, it means that the randomness of the operator is greater.

[0027] In some embodiments, determining the mapping relationship between operators and system resources based on the first log set and the second log set includes: training and acquiring a resource evaluation model based on the first log set and the second log set, the resource evaluation model including the mapping relationship between the operators and system resources, the resource evaluation type being used to receive the input number of operators of each resource occupancy type and output the occupancy of the system resources.

[0028] By classifying the multiple system operation logs included in the first log set, operators for implementing various operations performed by the network management system are identified, and based on the resource monitoring logs included in the second log set, the resource occupancy type corresponding to each operator in the first operator set is further identified, and the resource occupancy type is a valid operator, a periodic operator, or an invalid operator. Based on the number of operators of each resource occupancy type in the second log set and the first operator set, the mapping relationship between the operator and the system resource is determined, thereby realizing the identification and positioning of system resource consumption at the operator level, improving the accuracy of dynamic evaluation and prediction of system resource consumption, and thus improving the accuracy of resource scheduling.

[0029] In some embodiments, based on the second target time period, a third log set can be obtained from the first log set, and a fourth log set can be obtained from the second log set. The third log set can include at least part of the system operation logs in the first log set, and the fourth log set can include at least part of the resource monitoring logs in the second log set. Based on the third log set and the fourth log set, a sample set is generated, wherein the sample set includes a plurality of samples, each sample includes the number of operators of each resource occupancy type, and each sample is labeled with the actual system resource occupancy. The sample is input into a resource evaluation model, and the predicted system resource occupancy output by the resource evaluation model is obtained. Based on the difference between the predicted system resource occupancy and the actual system resource occupancy, the initial parameters in the resource evaluation model (such as the weight values ​​corresponding to the neurons mentioned above) are iteratively updated.

[0030] In some implementations, the resource assessment model includes a multi-layer perceptron model.

[0031] In some embodiments, the scheduling of system resources of the network management system based on the mapping relationship between the operator and the system resources includes: determining the occupancy of the system resources at a first moment based on the mapping relationship between the operator and the system resources; determining the function deployment strategy of the network management system at the first moment based on the occupancy of the system resources at the first moment and the function deployment strategy of the network management system at the second moment, the first moment being later than the second moment; and scheduling the system resources based on the function deployment strategy of the network management system at the first moment.

[0032] Based on the mapping relationship between operators and system resources, the occupation of system resources at the first moment can be determined, based on the occupation of system resources at the first moment and the function deployment strategy of the network management system at the second moment, the function deployment strategy of the network management system at the first moment can be determined, and based on the function deployment strategy of the network management system at the first moment, the system resources can be scheduled. That is, the identification results of operators and system resources can be fully utilized to achieve optimal configuration and acceleration of system resources, thereby improving user experience.

[0033] In a second aspect, an embodiment of the present application provides a management device based on a network management system, including:

[0034] An acquisition module, configured to acquire a first log set and a second log set from a network management system, wherein the first log set includes a plurality of system operation logs, and the second log set includes at least one resource monitoring log;

[0035] A determination module, configured to determine a mapping relationship between an operator and a system resource based on the first log set and the second log set, wherein the operator is used to implement an operation performed by the network management system;

[0036] The scheduling module is used to schedule the system resources of the network management system based on the mapping relationship between the operator and the system resources.

[0037] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory and a processor, the memory being used to store a computer program; the processor being used to execute any one of the methods described in the first aspect above when calling the computer program.

[0038] In a fourth aspect, an embodiment of the present application provides a chip system, which includes a processor, the processor is coupled to a memory, and the processor executes a computer program stored in the memory to implement any method described in the first aspect above.

[0039] The chip system may be a single chip or a chip module composed of multiple chips.

[0040] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method described in any one of the above-mentioned first aspects is implemented.

[0041] In a sixth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes any one of the methods described in the first aspect above.

[0042] It can be understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the system architecture of a network management system provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of the system architecture of another network management system provided in an embodiment of the present application;

[0046] Figure 4 A flowchart of a management method for a network management system provided by an embodiment of the present application;

[0047] Figure 5A flowchart of another management method of a network management system provided by an embodiment of the present application;

[0048] Figure 6 A flowchart of a method for determining a mapping relationship between an operator and system resources provided in an embodiment of the present application;

[0049] Figure 7 A schematic diagram of the structure of a multi-layer perceptron model provided in an embodiment of the present application;

[0050] Figure 8 A schematic diagram of an overfitting value curve of a multilayer perceptron model provided in an embodiment of the present application;

[0051] Fig. 9 A schematic diagram of the prediction results of a multi-layer perceptron model provided in an embodiment of the present application;

[0052] Fig.10 A flowchart of a method for scheduling system resources provided in an embodiment of the present application;

[0053] Fig.11 A flowchart of another method for scheduling system resources provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The management method based on the network management system provided in the embodiment of the present application can be applied to electronic devices such as computers. The embodiment of the present application does not impose any restrictions on the specific type of electronic devices.

[0055] Figure 1 1 is a schematic diagram of the structure of an electronic device 100 provided in an embodiment of the present application. The electronic device 100 may include a processor 110, a memory 120, a communication module 130, and the like.

[0056] The processor 110 may include one or more processing units, and the memory 120 is used to store program codes and data. In the embodiment of the present application, the processor 110 may execute computer-executable instructions stored in the memory 120 to control and manage the actions of the electronic device 100.

[0057] The communication module 130 can be used for communication between various internal modules of the electronic device 100, or for communication between the electronic device 100 and other external electronic devices. For example, if the electronic device 100 communicates with other electronic devices via a wired connection, the communication module 130 may include an interface, such as a USB interface. The USB interface may be an interface that complies with USB standard specifications, specifically a Mini USB interface, a Micro USB interface, a USB Type C interface, etc. The USB interface may be used to connect a charger to charge the electronic device 100, or may be used to transmit data between the electronic device 100 and peripheral devices. For example, the communication module 130 may be used for communication between the electronic device 100 and other devices in the OSS.

[0058] Optionally, the electronic device 100 may further include a display screen 140, and the display screen 140 may display images or videos in a human-computer interaction interface, etc. Exemplarily, the display screen 140 may display an OSS-related management interface.

[0059] Optionally, the electronic device 100 may further include a peripheral device 150 , such as a mouse, a keyboard, a speaker, a microphone, etc.

[0060] It should be understood that, in addition to Figure 1 In addition to the various components or modules listed in the figure, the embodiment of the present application does not specifically limit the structure of the electronic device 100. In other embodiments of the present application, the electronic device 100 may also include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure can be implemented in hardware, software, or a combination of software and hardware.

[0061] In order to facilitate understanding of the technical solutions in the embodiments of the present application, the application scenarios of the embodiments of the present application are first introduced below.

[0062] OSS is a network management system used in the telecommunications field and is a necessary support platform for the development and operation of telecommunications services. OSS can be responsible for the inspection and management of the communication quality and operation of the telecommunications network, record and collect various data in the operation of the entire network, and perform monitoring and control functions on various devices (such as base stations, etc.). The increasing scale and new features of telecommunications networks have also brought new challenges to the system capabilities of OSS. For example, the data volume of 5G networks has increased by dozens of times compared to 4G networks, which requires the corresponding OSS management capabilities and scale to increase by dozens of times. For example, OSS may include hundreds of high-performance servers, thousands of operating systems, hundreds of thousands of processes, millions or even tens of millions of processes and handles. Larger and more complex OSSs require more effective management methods, including resource evaluation and optimization of OSS systems.

[0063] It is understandable that, in practical applications, the network management system is not limited to OSS. For example, the network management system may also be a cluster management system, which may be used to manage a cluster of devices such as computers.

[0064] In some embodiments, the network management system can be deployed in Figure 1 The electronic device 100 shown in the figure may also be deployed in a distributed manner on a device cluster consisting of multiple electronic devices 100.

[0065] In some implementations, the electronic device 100 may be referred to as a physical machine. The network management system may include multiple virtual machines, which may be deployed on one or more physical machines.

[0066] Please refer to Figure 2 , is a schematic diagram of the system architecture of a network management system provided in an embodiment of the present application. The network management system includes a device access system, a data processing system, and a forwarding and presentation system.

[0067] The device access system can be used to connect devices in the target network to be managed to the network management system. For example, in OSS, the device access system can connect network elements such as base stations and controllers in the telecommunications network to OSS. For example, the device access system can be used to obtain data from the connected device and issue control instructions to the device. Of course, the device access system can also be used to perform other services related to device access.

[0068] The data processing system can be used to process the data reported by the devices in the target network, so as to realize the device management services such as device monitoring, device tracking and device alarm.

[0069] The forwarding and presentation system can be used to present a user interface to the user, thereby presenting data in the network management system to the user, receiving user operations, etc. And / or, the forwarding and presentation system can realize communication between the network management system and other external devices, such as forwarding data to other external devices.

[0070] In some embodiments, the network management system further includes an information collection system, which can collect system information from various systems included in the network management system and present the collected system information through an information collection interface. In some embodiments, the information collection process and collection results can be presented to the user through the information collection interface, so that the user can configure the information collection system, such as configuring the system to be collected, the type of data to be collected, the collection period, etc.

[0071] To Figure 2Take the network management system shown in the figure as an example. In some implementations, the user can obtain the system information of each system in the network management system through the information collection system, analyze the system information presented by the information collection system based on management experience, evaluate the resource consumption of the network management system, and adjust the parameters of each system so that each system can also run according to the new parameters. However, this management method requires users to learn and accumulate management experience for a long time, is highly subjective, and has poor management accuracy. In particular, as the scale of the target network continues to expand, it is impossible to achieve comprehensive management of the network management system by manual experience alone.

[0072] In order to solve at least part of the above technical problems, the present application embodiment provides another network management system. Figure 3 , the network management system Figure 2 The system shown also includes an analysis system and an acceleration system.

[0073] The information collection system can be used to collect system operation logs and resource monitoring logs in the network management system. System operation logs and resource monitoring logs can be generated by the network management system during operation. System operation logs, also known as operation logs, can record operations performed by the network management system. For example, taking the alarm service in OSS as an example, the system operation log can record various types of operations such as obtaining alarm events, warehousing alarm events, parsing alarm events, and formatting. Resource monitoring logs can be used to record the usage of system resources in the network management system, such as processor resources, memory resources, and disk resources, and it can be understood that in actual applications, resource monitoring logs can also record the usage of more or fewer types of resources. It can also be understood that in actual applications, system operation logs and resource monitoring logs can have other names.

[0074] The analysis system can be used to analyze the system operation log and the resource monitoring log to realize the evaluation and prediction of the system resources of the network management system. In some embodiments, the analysis system can classify the system operation log so that each type of system operation log corresponds to an operator, wherein the operator can refer to a functional unit for realizing a certain operation or a certain specific data analysis target, and an operator corresponds to a type of operation. In some embodiments, a process can be used to realize a service, and a service can include multiple operators. For example, the alarm service mentioned above can correspond to a process, and the process includes operators corresponding to multiple operations such as obtaining alarm events, storing alarm events, parsing alarm events, and formatting processing. During the operation of the process, a corresponding system operation log can be generated. Accordingly, the system operation log can be analyzed to obtain operators corresponding to multiple operations such as obtaining alarm events, storing alarm events, parsing alarm events, and formatting processing. In some embodiments, the analysis system can combine the classified operators and the resource monitoring log to determine the use of system resources by each operator, establish a mapping relationship between operators and system resources, and then realize the evaluation and prediction of the system resources. In some embodiments, the analysis system can be used to implement the analysis of the dependency relationship between various parameters in the network management system, the analysis of the dependency relationship between various system resources, or the analysis of the dependency relationship between various operators. In some embodiments, the analysis process and results of the analysis system can be presented to the user through the analysis interface, so that the user can configure the acceleration system.

[0075] The acceleration system can be used to establish a resource scheduling model based on the evaluation and prediction results of the analysis system, realize the resource optimization configuration of the network management system, and improve the user experience. In some implementations, the acceleration process, acceleration strategy or acceleration result can be presented to the user through the system acceleration interface.

[0076] It is understandable that the embodiments of the present application are only Figure 2 and Figure 3 , a possible system architecture of a network management system is described, rather than limiting the network management system. In practical applications, the network management system can implement a set of functions configured by a user or a developer, and the set of functions can include one or more functions. The network management system can be a multi-layer structure, and each layer of the structure can implement at least part of the functions in the set of functions. Exemplarily, in order from high level to low level, the network management system can include a system level, a subsystem level, a module level, a service level, and an operator level, wherein each level can include one or more higher levels. The operator level can include multiple operators, and the operator can serve as a functional unit in the network management system. In some embodiments, the function that an operator can implement can include only one operation.

[0077] For example, in Figure 3 Take the network management system shown as an example. The network management system includes five subsystems, namely, the device access system, the data processing system, the forwarding and presentation system, the analysis system and the acceleration system. The device access system, the data processing system, the forwarding and presentation system are respectively composed of three modules, namely, information collection, business processing and parameter adjustment. The information collection module and the business processing module may include an alarm service. The alarm service may include operators that respectively implement various types of operations, such as obtaining alarm events, warehousing alarm events, parsing alarm events, and formatting processing. That is, the operator can be used as the smallest functional unit in the network management system, and then as the smallest measurement unit of occupied resources.

[0078] Please continue to refer to Figure 4 , is a flow chart of a management method of a network management system provided in an embodiment of the present application, which method can be used in the aforementioned network management system. Figure 4 It can be seen that in the embodiment of the present application, the system logs in the network management system can be obtained, including the system operation logs and resource monitoring logs. The system operation logs are analyzed to obtain multiple operators, and the mapping relationship between the operators and the system resources is constructed in combination with the resource monitoring logs, and then the evaluation and prediction of the occupation of the system resources by the operators are determined, and then the optimization configuration strategy for the system resources is obtained based on the evaluation and prediction results, and the system resources of the network management system are scheduled and configured based on the optimization configuration strategy, thereby realizing the automatic analysis and optimization of the system resources, and significantly improving the management capabilities and user experience of the network management system.

[0079] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0080] Please refer to Figure 5 , is a flow chart of a management method of a network management system provided by an embodiment of the present application. It should be noted that the method is not based on Figure 5 The specific order described below is a limitation, and it should be understood that in other embodiments, the order of some steps in the method can be interchanged according to actual needs, or some steps can be omitted or deleted. The method includes the following steps:

[0081] S501, the network management system obtains a first log set and a second log set from the network management system, the first log set includes a plurality of system operation logs, and the second log set includes at least one resource monitoring log.

[0082] In some implementations, the resource monitoring log may be used to record the usage of system resources. For example, the system resources may include one or more of central processing unit (CPU) resources, memory resources, hard disk resources, or input / output (I / O) resources.

[0083] In some implementations, in order to improve the accuracy of obtaining system operation logs and resource monitoring logs, at least one of a first target time period, a target system range, and a target resource type may be determined, and a first log set and a second log set may be obtained from a network management system based on at least one of the first target time period, the target system range, and the target resource type. In some implementations, at least one of the first target time period, the target system range, and the target resource type submitted by a user may be received. Of course, in actual applications, at least one of the first target time period, the target system range, and the target resource type may also be determined by other means.

[0084] The first target time period can be used to indicate the time period for obtaining system operation logs and resource monitoring logs. Based on the first target time period, the system operation logs and resource monitoring logs belonging to the first target time period can be obtained. For example, the first target time period may include all time periods before the current moment, within the last 1 day, within a certain day specified in the history, and the like.

[0085] The target system scope can be used to indicate at least one system included in the network management system, that is, to indicate the source of the system operation log and resource monitoring log. Based on the target system scope, the system operation log and resource monitoring log derived from the target system scope can be obtained. For example, the target system scope can include at least one of a device access system, a data processing system, and a forwarding and presentation system.

[0086] The target resource type may be used to indicate the type of system resource recorded in the acquired resource monitoring log.

[0087] For example, in Figure 3 In the network management system shown, the first target time period, target system range and target resource type submitted by the user can be received through the information collection interface. Based on the first target time period, target system range and target resource type and other information, the information collection system obtains the system operation log and resource monitoring log from at least one of the device access system, data processing system and forwarding and presentation system indicated by the target system range.

[0088] S502: The network management system determines a mapping relationship between an operator and system resources based on the first log set and the second log set, where the operator is used to implement an operation performed by the network management system.

[0089] From the above, it can be seen that the system operation log included in the first log set and the resource monitoring log included in the second log set can be generated by the network management system during the operation process, wherein the system operation log can indicate the various operations performed by the network management system, and the resource monitoring log can indicate the occupancy of each system resource. Therefore, based on the first log set and the second log set, it is possible to determine the system resources occupied by the operators that perform various operations of the network management system, that is, to determine the mapping relationship between the operators and the system resources.

[0090] It is understandable that the embodiments of the present application do not limit the specific relationship structure of the mapping relationship between operators and system resources. Exemplarily, the mapping relationship between operators and system resources may include one or more operators and one or more system resources occupied by each operator. Exemplarily, the mapping relationship between operators and system resources may include one or more system resources and operators occupying each system resource.

[0091] It can also be understood that the embodiment of the present application does not limit the storage form of the mapping relationship between the operator and the system resource. Exemplarily, the mapping relationship between the operator and the system resource can be stored in the form of a data table. Exemplarily, the mapping relationship between the operator and the system resource can be stored in the form of a machine learning model.

[0092] For example, in Figure 3 In the network management system shown, the first log set and the second log set can be analyzed by the analysis system to determine the mapping relationship between the operator and the system resources. The analysis process and analysis results can also be displayed to the user through the analysis interface.

[0093] S503: The network management system schedules system resources of the network management system based on the mapping relationship between the operator and the system resources.

[0094] Since the mapping relationship between operators and system resources is determined, the system resources occupied by the operators of the network management system to perform various operations are also determined. Therefore, if the operations performed by the network management system are determined, the system resources required for the network management system to perform these operations can be determined; conversely, if the system resources occupied by the network management are determined, the operations performed by the network management system can be determined, and then the system resources of the network management system can be scheduled. For example, when the system resources required for the network management system to perform a certain operation may be insufficient, more system resources are allocated to the operation in a timely manner; when the system resources required for the network management system to perform a certain operation may be redundant, idle system resources are recovered in a timely manner, and so on. That is, based on the system logs of the network management system such as system operation logs and resource monitoring logs, it is possible to realize automatic analysis and optimization of system resources, significantly improving the management capabilities and user experience of the network management system.

[0095] For example, in Figure 3 In the network management system shown, a specific scheduling strategy can be analyzed and determined based on the mapping relationship between operators and system resources by analyzing the system, and the scheduling strategy can be executed by the acceleration system. The scheduling strategy, the scheduling process and results of system resources can also be displayed to the user through the acceleration interface.

[0096] In an embodiment of the present application, a first log set and a second log set in a network management system can be obtained, the first log set includes multiple system operation logs, and the second log set includes at least one resource monitoring log. Among them, the system operation log can record the operations performed by the network management system, and the resource monitoring log can record the occupancy of the system resources of the network management system. Therefore, based on the system operation log and the resource monitoring log, a mapping relationship between operators that perform various operations and system resources can be constructed. Based on this mapping relationship, the demand for system resources when the network management system performs various operations can be determined, and therefore, based on the mapping relationship between the operator and the system resources, the system resources of the network management system can be scheduled, thereby realizing automatic analysis and optimization of system resources, significantly improving the management capabilities and user experience of the network management system.

[0097] Please refer to Figure 6 , is a flow chart of a method for determining a mapping relationship between an operator and a system resource provided in an embodiment of the present application. The method can be used as a detailed description of the aforementioned S502. It should be noted that the method is not based on Figure 6 The specific order described below is a limitation, and it should be understood that in other embodiments, the order of some steps in the method can be interchanged according to actual needs, or some steps can be omitted or deleted. The method includes the following steps:

[0098] S601: The network management system pre-processes a plurality of system operation logs in a first log set.

[0099] Preprocessing the system operation logs can also be called cleaning the system operation logs. Preprocessing can reduce the number of system operation logs that cannot be applied to subsequent steps (ie, invalid system operation logs), and improve the accuracy of subsequent operator identification and classification.

[0100] In some embodiments, the preset characters included in the system operation log may be deleted without changing the order of other characters in the system operation log. In some embodiments, the preset characters may include invalid characters, erroneous characters, or blank characters. In some embodiments, the system operation log may be analyzed, such as semantic analysis, to identify invalid characters and erroneous characters in the system operation log.

[0101] It is understandable that in practical applications, the system logs may also be preprocessed in other ways.

[0102] In some implementations, S601 may be omitted.

[0103] S602: The network management system classifies a plurality of system operation logs included in a first log set to obtain a first operator set, where the first operator set includes at least one operator.

[0104] Each operator is used to implement a type of operation performed by the network management system, and each operator corresponds to at least part of the system operation logs in the first log set.

[0105] In some embodiments, the system operation logs may be regarded as strings, so that the multiple system operation logs included in the first set may be classified according to the classification method of the strings. In some embodiments, the multiple system operation logs may be classified based on the string lengths and / or similarities of the multiple system operation logs included in the first log set to obtain a first operator set, wherein the difference between the string lengths of the multiple system operation logs corresponding to the same operator is less than a first value, and the similarity of the multiple system operation logs corresponding to the same operator is greater than or equal to a second value. That is, the lengths of the multiple system operation logs corresponding to the same operator are the same or similar, and the strings included in the multiple system operation logs are the same or similar.

[0106] Exemplarily, the multiple system operation logs in the first set may be partitioned based on the string lengths of the multiple system operation logs to obtain at least one partition, each partition including at least part of the system operation logs in the first set, and the string lengths of the system operation logs included in each partition are the same or similar, and then the system operation logs in each partition may be classified based on the similarity of the system operation logs in each partition to obtain at least one operator, each operator corresponding to at least part of the system operation logs in the first set. Among them, since the possibility of system operation logs with large length differences being recorded as the same type of operation (i.e. corresponding to the same operator) is small, the system operation logs in the first set may be preliminarily classified based on the lengths of the system operation logs, which can significantly improve the efficiency of classifying the system operation logs in the first set.

[0107] In some embodiments, partitioning the multiple system operation logs based on the string lengths of the multiple system operation logs in the first set may include: creating at least one partition, each partition corresponding to a string length, and determining the string lengths of each system operation log in the first set. If the difference between the string length of the system operation log and the string length corresponding to any partition is less than a first value, then determining that the system operation log belongs to any partition. If the difference between the string length of the system operation log and the string length corresponding to each partition is greater than or equal to the first value, then creating a new partition, and determining that the new system operation log belongs to the new partition, and the new partition corresponds to the string length of the system log.

[0108] In some implementations, the similarity between any two system operation logs may be determined based on the longest common subsequence. It is understandable that in practical applications, the similarity may also be determined in other ways.

[0109] It should be noted that the first value and the second value can be obtained by prior setting, and it can be understood that the embodiment of the present application does not limit the method of setting the first value and the second value and the numerical values ​​of the first value and the second value.

[0110] It is understandable that, in actual applications, the multiple system operation logs included in the first set may also be classified according to other character string classification methods.

[0111] In some implementations, system operation logs with a number of occurrences greater than or equal to a third value may be obtained from at least part of the system operation logs corresponding to the operator, and the operator may be labeled based on the system operation logs with a number of occurrences greater than or equal to the third value. The system operation logs with a number of occurrences greater than or equal to the third value may be the most or more system operation logs corresponding to the operator, and the operator may be labeled based on the system operation logs with a number of occurrences greater than the third value, that is, the most or more system operation logs corresponding to the operator are used as the label of the operator, so as to facilitate the determination of the specific operation and other features corresponding to the operator, and further distinguish and identify different operators.

[0112] It should be noted that the third value can be obtained by prior setting, and it can be understood that the embodiment of the present application does not limit the method of setting the third value and the value of the third value.

[0113] It is understandable that in practical applications, operators may also be labeled in other ways.

[0114] S603: The network management system determines a resource occupancy type corresponding to each operator in the first operator set based on the second log set, where the resource occupancy type is a valid operator, a periodic operator, or an invalid operator.

[0115] In some implementations, valid operators correspond to data analysis type operations, and the number of valid operators has the strongest correlation with the amount of system resources occupied, and the more valid operators there are, the greater the amount of system resources occupied; periodic operators correspond to scheduled task type operations, and the number of periodic operators has a low correlation with the amount of system resources occupied, and the increase or decrease of periodic operators will not cause a significant change in the amount of system resources occupied; invalid operators correspond to log printing type operations, and the number of invalid operators has the lowest correlation with the amount of system resources. When the resource occupation type corresponding to each operator in the first operator set is determined, the amount of system resources that each operator in the first operator set may occupy is also determined.

[0116] It is understandable that in practical applications, the valid operator, the periodic operator and the invalid operator may also have other names, and the embodiments of the present application do not limit the names of these three resource occupation types. For example, the invalid operator may also be called a basic operator.

[0117] In some embodiments, based on the second log set, the fluctuation characteristics of the system resource occupancy in multiple time periods and the distribution characteristics of the operators in multiple time periods can be determined; based on the fluctuation characteristics of the system resource occupancy in multiple time periods and the distribution characteristics of the operators in multiple time periods, the resource occupancy type corresponding to the operator can be determined.

[0118] In some embodiments, the total time period of the first log set and the second log set can be obtained, and the total time period can also be called a monitoring period. The total time period is divided into multiple time periods, and the duration of each time period can be the first duration. Based on the first log set and the multiple time periods, the operators included in the network management system in each time period can be determined. By comparing the operators included in the network management system in the multiple time periods, the distribution characteristics of the operators of the network management system in the multiple time periods can be determined, such as whether each operator appears and the frequency of occurrence. Based on the second log set and the multiple time periods, the occupation of system resources of the network management system in each time period can be determined. By comparing the occupation of system resources of the network management system in the multiple time periods, the distribution characteristics of the system resources of the network management system in the multiple time periods can be determined, such as the change of the occupation of system resources. Based on the degree of synchronization between the time period from the appearance to the disappearance of the operator and the time period from the increase to the decrease of the system resources, the resource occupation type of the operator is determined. Exemplarily, during the period of system resource change, the operators that also appear densely are valid operators, the operators whose number remains basically unchanged are periodic operators, and the operators that appear randomly are invalid operators.

[0119] Exemplarily, a monitoring cycle includes 3 consecutive time periods, each of which is 30 seconds long (i.e., the first time period). In the first to third time periods, the CPU occupancy of the network management system is 5%, 8%, and 5%, respectively. Operator 1 does not appear in the first time period, operator 1 appears in the second time period, and operator 1 does not appear in the third time period. Since in the second time period, operator 1 appears and the CPU occupancy increases, and in the first and third time periods, operator 1 disappears and the CPU occupancy decreases, it can be determined that operator 1 is a valid operator.

[0120] Alternatively, in the first to third time periods, the CPU usage of the network management system is 5%, and operator 1 appears or disappears randomly. Since the appearance and disappearance of operator 1 does not cause a change in the CPU usage, operator 1 can be determined to be an invalid operator.

[0121] Alternatively, in the first to third time periods, the CPU occupancy of the network management system is 5%, 5%, and 3% respectively. Operator 1 appears in the first and second time periods, and the frequency of operator 1 appearing in the second time period is higher than the frequency of appearing in the first time period. Operator 1 does not appear in the third time period. Since the CPU occupancy increases when operator 1 appears in the first and second time periods, but the change in the frequency of operator 1 does not affect the change in the CPU occupancy, it can be determined that operator 1 is a periodic operator.

[0122] It should be noted that the first duration can be obtained through pre-setting, and it can be understood that the embodiment of the present application does not limit the method of setting the first duration and the numerical value of the first duration.

[0123] In some embodiments, the entropy value corresponding to the operator can be determined, and based on the entropy value corresponding to the operator, the resource occupation type corresponding to the operator is verified, so that the classification result according to the entropy value is cross-validated with the classification result according to the resource monitoring log, so as to improve the accuracy of identifying and classifying the operator. In some embodiments, when the entropy value of the operator is less than the fourth value (for example, the entropy value is a smaller value such as 0), the resource occupation type of the operator is determined to be a periodic operator; when the entropy value of the operator is greater than or equal to the fourth value and less than the fifth value (that is, the entropy value is a medium value), the resource occupation type of the operator is determined to be a periodic operator; when the entropy value of the operator is greater than or equal to the fifth value (that is, the entropy value is a larger or maximum value), the resource occupation type of the operator is determined to be an invalid operator. Among them, the entropy value corresponding to the operator is used to indicate the randomness of the distribution of the operator in multiple time periods. When the entropy value is larger, it means that the randomness of the operator is smaller; when the entropy value is smaller, it means that the randomness of the operator is larger.

[0124] It should be noted that the fourth value and the fifth value can be obtained by prior setting, and it can be understood that the embodiment of the present application does not limit the method of setting the fourth value and the fifth value and the numerical value size of the fourth value and the fifth value.

[0125] For example, if the monitoring period is T (minutes), the monitoring period is divided into T time periods, each time period is 1 minute, and the operator matrix within the monitoring period can be constructed as shown in the following formula 1.

[0126] R=(x ij )mn, (i=1, 2, ..., n; j=1, 2, ..., T) (Formula 1)

[0127] Where R represents the operator matrix, x ij It represents the number of operators of the i-th type appearing at time j, n represents the type of operator, and T represents the T-th time period.

[0128] By normalizing the operator matrix, a normalized matrix B can be obtained. Each element in the normalized matrix B can be expressed as in the following formula 2.

[0129]

[0130] Among them, b ij The number of operators of the i-th type at the j-th time is the normalized parameter value, x max and x minRespectively represent the maximum and minimum values ​​of the number of operators of the same type in this time period.

[0131] According to the definition of entropy, the entropy value H of the i-th type operator in m time periods i It can be shown as the following formula 3.

[0132]

[0133] S604: The network management system determines a mapping relationship between operators and system resources based on the second log set and the number of operators of each resource occupancy type in the first operator set.

[0134] In some embodiments, a resource evaluation model can be trained based on the first log set and the second log set. The resource evaluation model includes a mapping relationship between operators and system resources. The resource evaluation type is used to receive the input number of operators of each resource occupancy type and output the occupancy of system resources.

[0135] In some embodiments, the resource evaluation model may include a multi-layer perceptron model. Figure 7 As shown in the figure, the multilayer perceptron model may include an input layer, multiple hidden layers, and an output layer. The input layer can receive input features. The hidden layer includes multiple neurons that can extract features and perform nonlinear transformations. Each hidden layer is fully connected to the previous layer, receives the output of the previous layer, performs weighted and activation function operations, and obtains the output of the current layer. The output layer can be a fully connected layer, which is used to give the final prediction result. Through iterative training, the multilayer perceptron model can learn the complex relationship between input features and predict new data. For example, in Figure 7 In the multilayer perceptron model shown, the multiple hidden layers include hidden layer 1, hidden layer 2 and hidden layer 3, wherein hidden layer 1 includes 100 neurons, hidden layer 2 includes 50 neurons, hidden layer 3 includes 20 neurons, and the output layer includes 1 neuron. Each neuron can correspond to a weight value. By training the multilayer perceptron model, the weight values ​​corresponding to these neurons can be updated, and ultimately the predicted value output by the multilayer perceptron model is made the same or close to the true value.

[0136] In some embodiments, based on the second target time period, a third log set can be obtained from the first log set, and a fourth log set can be obtained from the second log set. The third log set may include at least part of the system operation logs in the first log set, and the fourth log set may include at least part of the resource monitoring logs in the second log set. Based on the third log set and the fourth log set, a sample set is generated, wherein the sample set includes a plurality of samples, each sample includes the number of operators of each resource occupancy type, and each sample is labeled with the actual system resource occupancy. The sample is input into a resource evaluation model, and the predicted system resource occupancy output by the resource evaluation model is obtained. Based on the difference between the predicted system resource occupancy and the actual system resource occupancy, the initial parameters in the resource evaluation model (such as the weight values ​​corresponding to the neurons mentioned above) are iteratively updated. Alternatively, in other embodiments, a sample set can be generated directly based on the first log set and the second log set.

[0137] In some embodiments, the sample set can be divided into a training set and a test set, and the training set and the test set can include some samples in the sample set. The resource evaluation model is obtained by training the training set, and the obtained resource evaluation model is tested by the test set. If the prediction result of the resource evaluation model for the test set is greater than or equal to a preset accuracy threshold, the training can be stopped. If the prediction result of the resource evaluation model for the test set is less than the preset accuracy threshold, the resource evaluation model can continue to be trained.

[0138] Exemplarily, the duration of the second target time period may be greater than or equal to 300 minutes and less than 480 minutes. Exemplarily, the training set may include 70% of the samples in the sample set, and the test set may include 30% of the samples in the sample set. It is understandable that, in actual applications, the second target time period may also be a time period of other lengths, and the sample set may also be divided into a training set and a test set according to other proportions.

[0139] Taking the multi-layer perceptron model as an example, as the training time increases, the change trend of the overfitting value R2 of the multi-layer perceptron model can be shown as follows: Figure 8 As shown. The overfitting value of the multilayer perceptron model can refer to the gap between the training error and the test error. The closer the overfitting value is to 1, the better the fitting of the multilayer perceptron model is, that is, the smaller the gap between the training error and the test error is. Figure 8 The lower right corner shows the process to which the operator input to the multi-layer perceptron model belongs. Figure 8 It can be seen that when the training time is greater than or equal to 360 minutes, the overfitting value of the multilayer perceptron model tends to be stable. The CPU usage is predicted by the multilayer perceptron model, and the following is obtained: Fig. 9 The prediction results are shown in Figure 2. Fig. 9 It can be seen that the predicted value of the CPU usage by the multilayer perceptron model has a high degree of overlap with the actual value, which means that the accuracy of the multilayer perceptron model is very high.

[0140] In some embodiments, in order to ensure the accuracy of the resource assessment model, after the resource assessment model is trained, the sample set can be updated once or multiple times (for example, periodically) based on the newly acquired system operation logs and resource monitoring logs, and the resource assessment model can be trained again based on the updated sample set.

[0141] In an embodiment of the present application, the multiple system operation logs included in the first log set are classified to identify operators for implementing various operations performed by the network management system, and based on the resource monitoring logs included in the second log set, the resource occupancy type corresponding to each operator in the first operator set is further identified, and the resource occupancy type is a valid operator, a periodic operator, or an invalid operator. Based on the number of operators of each resource occupancy type in the second log set and the first operator set, the mapping relationship between the operator and the system resource is determined, thereby realizing the identification and positioning of system resource consumption at the operator level, improving the accuracy of dynamic evaluation and prediction of system resource consumption, and thus improving the accuracy of resource scheduling.

[0142] Please refer to Fig.10 , is a flow chart of a method for scheduling system resources provided in an embodiment of the present application. The method can be used as a detailed description of the aforementioned S503. It should be noted that the method is not based on Fig.10 The specific order described below is a limitation, and it should be understood that in other embodiments, the order of some steps in the method can be interchanged according to actual needs, or some steps can be omitted or deleted. The method includes the following steps:

[0143] S1001, the network management system determines the occupation of system resources at the first moment based on the mapping relationship between operators and system resources.

[0144] In some implementations, the first moment may be any moment later than the current moment. Of course, in practical applications, the first moment may also be other moments.

[0145] In some implementations, the occupation amount of the system resources of the network management system at the first moment can be determined based on the second operator set included in the network management system at the first moment and the mapping relationship between the operator and the system resources. The second operator set may include at least one operator. In some implementations, the number of operators of each resource occupation type in the second operator set may be input into the aforementioned resource evaluation model, and the occupation amount of the resource system at the first moment output by the resource evaluation model is obtained.

[0146] In some embodiments, the function set included in the network management system at the first moment can be determined, and the operations included in the function set can be split to obtain a second operator set, wherein the function set can include services or processes. Alternatively, in other embodiments, the function set included in the network management system before the first moment can be determined, and the operations included in the function set can be split to obtain a third operator set, and the second operator set can be predicted based on the third operator set. Alternatively, in other embodiments, the function set included in the network management system at the first moment can be predicted based on the function set included in the network management system before the first moment, and then the second operator set can be determined based on the function set included in the network management system at the first moment. Of course, in practical applications, the second operator set can also be determined in other ways.

[0147] S1002, the network management system determines a function deployment strategy of the network management system at the first moment based on the occupation of system resources at the first moment and the function deployment strategy of the network management system at the second moment, the first moment being later than the second moment.

[0148] The function deployment strategy may include the deployment location of the function set in the network management system, the system resources allocated to the function set, etc. For example, the network management system is a distributed system deployed on multiple physical machines, and the network management system includes multiple virtual machines, which may be deployed on some or all of the multiple physical machines, each virtual machine may include some services, some functions, and some processes in the network management system, and the function deployment strategy may include the deployment location of the multiple virtual machines.

[0149] In some implementations, a look-ahead rolling model may be used to determine the function deployment strategy of the network management system at the first moment. Exemplarily, the look-ahead rolling model may be shown in Formula 4 below.

[0150]

[0151] Among them, k represents a certain moment, such as the first moment; x k represents the set of functions included in the network management system at time k; Ω k represents the total amount of system resources at time k; u k Indicates the usage of a specific system resource at time k, such as the CPU usage; I(x k-1 , k-1) represents the function deployment strategy of the network management system at time k-1 (for example, the second time); I(x k , k) represents the function deployment strategy of the network management system at time k.

[0152] For the above formula 4, I(x k , k), we can also combine the idea of ​​dynamic programming and constraint variable separation to realize the separation of discrete and mixed constraint variables in constraint variables, and quickly determine I(x by constructing the convex hull feasible domain. k , k). It can be understood that, in practical applications, the above formula 4 can also be solved by other methods.

[0153] S1003, the network management system schedules system resources based on the function deployment strategy of the network management system at the first moment.

[0154] In some implementations, the system resources allocated to the operator can be configured and updated to achieve dynamic adjustment of the resources consumed by the operator, and then the system resources consumed by higher-level functions such as the service or process to which the operator belongs can be more accurately scheduled, thereby improving the utilization of system resources and the operating efficiency of the network management system. For example, the initial system resources configured by the network management system for operator 1 include 1% of the CPU occupancy and 2% of the memory occupancy, but after predicting the mapping relationship between the operator and the system resources, it is determined that operator 1 actually only needs to occupy 0.5% of the CPU and 1% of the memory to be implemented, so the system resources configured for operator 1 are updated to include 0.5% of the CPU occupancy and 1% of the memory occupancy.

[0155] In some implementations, based on the hierarchical division and deployment of each function in the function set of the network management system, the system resources of the network management system can be scheduled from multiple levels and dimensions such as service level, module level and data flow to achieve acceleration of the network management system and improve user experience. Exemplarily, one or more of the following scheduling methods may be included.

[0156] In a possible scheduling method, the occupation of system resources corresponding to the network management system can be determined according to the occupation of system resources corresponding to each operator included in the network management system. If the occupation of the system resources is greater than the actual allocated occupation, the excess system resources can be recycled. If the occupation of the system resources is less than the actual allocated occupation, you can request to add more system resources. For example, the network management system currently occupies 8 core CPUs and 100G memory. After prediction, the system resource requirements of the network management system will be 6 core CPUs and 50G memory at multiple consecutive moments in the future. Then the excess 2 core CPUs and 50G memory can be released for use by other functions to improve resource utilization.

[0157] In one possible scheduling method, for an instance of any function in a function set of a network management system, if the current number of instances cannot meet the processing requirements of the function, one or more instances implementing the function are added to accelerate the function.

[0158] In one possible scheduling method, for an instance of any function in the function set of the network management system, if the current number of instances cannot meet the processing requirements of the function, the system resources allocated to the instance are increased to accelerate the function.

[0159] In a possible scheduling method, for instances implementing any function in a function set of a network management system, if the current number of instances is higher than the processing requirement for satisfying the function, one or more instances implementing the function are terminated to release system resources.

[0160] In a possible scheduling method, based on the load status of multiple virtual machines, at least part of the functions of a virtual machine with a higher load may be migrated to other virtual machines with a lower load.

[0161] In a possible scheduling method, based on the load status of multiple physical machines, at least part of the functions of a physical machine with a higher load may be migrated to other physical machines with a lower load.

[0162] In a possible scheduling method, based on the data flows input to multiple virtual machines, the data flows allocated to the virtual machines with higher data flows may be reduced, and the data flows allocated to the virtual machines with lower data flows may be increased.

[0163] In a possible scheduling method, based on the data flows input to multiple physical machines, the data flows allocated to the physical machines with higher data flows may be reduced, and the data flows allocated to the physical machines with lower data flows may be increased.

[0164] In the embodiment of the present application, the occupation of system resources at the first moment can be determined based on the mapping relationship between operators and system resources, the function deployment strategy of the network management system at the first moment can be determined based on the occupation of system resources at the first moment and the function deployment strategy of the network management system at the second moment, and the system resources can be scheduled based on the function deployment strategy of the network management system at the first moment. That is, the identification results of operators and system resources can be fully utilized to achieve optimal configuration and acceleration of system resources, thereby improving user experience.

[0165] Please refer to Fig.11, is a flow chart of a method for scheduling system resources provided in an embodiment of the present application. It should be noted that the method is not limited to the specific order described below. It should be understood that in other embodiments, the order of some steps in the method can be interchanged according to actual needs, or some steps can be omitted or deleted. The method includes the following steps:

[0166] In the first stage, during the operation of the network management system, when the network management system is under various load conditions such as periodic load and burst load, the resource law description in the network management system is determined based on the occupancy of various system resources such as CPU, memory and disk in the network management system. The resource law description may include the mapping relationship between the operators and system resources mentioned above.

[0167] In the second stage, the network management system can determine the current or future load status. The resource scheduling engine optimizes the initial function deployment strategy based on the resource description rule and obtains the optimized function deployment strategy. The resource scheduling engine executes the function deployment strategy through any scheduling framework such as the K8S job scheduling framework and the Yarm job scheduling framework, thereby scheduling system resources.

[0168] Based on the same inventive concept, the embodiment of the present application also provides a management device based on a network management system. The management device includes an acquisition module, a determination module and a scheduling module.

[0169] An acquisition module, used to acquire a first log set and a second log set from a network management system, the first log set including a plurality of system operation logs, and the second log set including at least one resource monitoring log;

[0170] A determination module, used to determine a mapping relationship between an operator and a system resource based on the first log set and the second log set, the operator being used to implement an operation performed by the network management system;

[0171] The scheduling module is used to schedule the system resources of the network management system based on the mapping relationship between operators and system resources.

[0172] Among them, the management device can implement various operations performed by the aforementioned network management system and has the same technical effects, which will not be described here one by one.

[0173] Based on the same inventive concept, an embodiment of the present application further provides an electronic device. The electronic device comprises: a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the method described in the above method embodiment when calling the computer program.

[0174] The electronic device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0175] Based on the same inventive concept, an embodiment of the present application further provides a chip system, which includes a processor coupled to a memory, and the processor executes a computer program stored in the memory to implement the method described in the above method embodiment.

[0176] The chip system may be a single chip or a chip module composed of multiple chips.

[0177] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described in the above method embodiment is implemented.

[0178] The embodiment of the present application also provides a computer program product. When the computer program product is run on an electronic device, the electronic device implements the method described in the above method embodiment when executing the computer program product.

[0179] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may at least include: any entity or device that can carry the computer program code to the camera / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

[0180] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0181] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0182] In the embodiments provided in the present application, it should be understood that the disclosed devices / equipment and methods can be implemented in other ways. For example, the device / equipment embodiments described above are merely schematic, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0183] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0184] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0185] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0186] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0187] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0188] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A management method based on a network management system, characterized in that: include: Acquire a first log set and a second log set from a network management system, wherein the first log set includes a plurality of system operation logs, and the second log set includes at least one resource monitoring log; Determine, based on the first log set and the second log set, a mapping relationship between an operator and a system resource, the operator being used to implement an operation performed by the network management system; Based on the mapping relationship between the operator and the system resources, the system resources of the network management system are scheduled.

2. The method according to claim 1, characterized in that The determining, based on the first log set and the second log set, a mapping relationship between operators and system resources includes: Classifying the multiple system operation logs included in the first log set to obtain a first operator set, wherein the first operator set includes at least one operator, wherein each operator is used to implement a type of operation performed by the network management system, and each operator corresponds to at least part of the system operation logs in the first log set; Determine, based on the second log set, a resource occupation type corresponding to each operator in the first operator set, where the resource occupation type is a valid operator, a periodic operator, or an invalid operator, wherein the valid operator corresponds to an operation of a data analysis type, the periodic operator corresponds to an operation of a scheduled task type, and the invalid operator corresponds to an operation of a print log type; Based on the second log set and the number of operators of each resource occupancy type in the first operator set, a mapping relationship between the operators and system resources is determined.

3. The method according to claim 2, characterized in that The classifying the plurality of system operation logs included in the first log set to obtain a first operator set includes: Based on the string lengths and / or similarities of the multiple system operation logs included in the first log set, the multiple system operation logs are classified to obtain the first operator set, wherein the difference between the string lengths of the multiple system operation logs corresponding to the same operator is less than a first value, and the similarity of the multiple system operation logs corresponding to the same operator is greater than or equal to a second value.

4. The method according to claim 2 or 3, characterized in that: Before classifying the plurality of system operation logs included in the first log set to obtain a first operator set, the method further includes: Delete the preset characters included in each of the system operation logs.

5. The method according to any one of claims 2 to 4, characterized in that: The method further comprises: Obtaining, from at least part of the system operation logs corresponding to the operator, system operation logs whose occurrence times are greater than or equal to a third value; The operator is marked based on the system operation log whose occurrence number is greater than or equal to the third value.

6. The method according to any one of claims 2 to 5, characterized in that: The determining, based on the second log set, the resource occupation type corresponding to each operator in the first operator set includes: Based on the second log set, determining fluctuation characteristics of the usage of the system resources in multiple time periods; Determining distribution characteristics of the operator in the multiple time periods; The resource occupation type corresponding to the operator is determined based on the fluctuation characteristics of the occupation of the system resources in multiple time periods and the distribution characteristics of the operator in the multiple time periods.

7. The method according to claim 6, characterized in that The method further comprises: Determine an entropy value corresponding to the operator, where the entropy value is used to indicate the randomness of the distribution of the operator in the multiple time periods; Based on the entropy value corresponding to the operator, the resource occupation type corresponding to the operator is verified.

8. The method according to any one of claims 1 to 7, characterized in that: The determining, based on the first log set and the second log set, a mapping relationship between operators and system resources includes: Based on the first log set and the second log set, a resource evaluation model is trained and obtained, wherein the resource evaluation model includes a mapping relationship between the operator and the system resources, and the resource evaluation type is used to receive the input number of operators of each resource occupancy type and output the occupancy of the system resources.

9. The method according to claim 8, characterized in that The resource evaluation model includes a multi-layer perceptron model.

10. The method according to any one of claims 1 to 9, characterized in that: The scheduling of the system resources of the network management system based on the mapping relationship between the operator and the system resources includes: Determine, based on a mapping relationship between the operator and the system resource, an occupation amount of the system resource at a first moment; Determine the function deployment strategy of the network management system at the first moment based on the occupation amount of the system resources at the first moment and the function deployment strategy of the network management system at the second moment, the first moment being later than the second moment; The system resources are scheduled based on the function deployment strategy of the network management system at the first moment.

11. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the method according to any one of claims 1 to 10 when calling the computer program.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

13. A computer program product, characterized in that When the computer program product is executed on an electronic device, the electronic device is enabled to execute the method according to any one of claims 1 to 10.

Images