Network optimization method and device

Abstract

A network optimization method and device, relating to the field of mobile communication technology, the method comprising: establishing a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value; predicting the temporal changes of the total end-to-end service value or the temporal changes of the end-to-end performance index of each optimization cluster according to the target mapping model, determining the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generating the optimization demand case of the optimization cluster; calculating and sorting the priority of the optimization demand cases of each optimization cluster, and optimizing each optimization demand case in turn according to the sorting result. The present invention can evaluate and sort the value of network optimization requirements, realize real-time monitoring of multi-dimensional network performance, and ensure service quality through the automatic generation and sorting of optimization requirements.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a network optimization method and device. Background Art

[0002] At present, in the fifth generation mobile communication (5G) wireless network, network optimization requirements are usually obtained based on the network management's monitoring and analysis of network performance indicators. For example, taking the daily network optimization workflow of a provincial company of a certain operator as an example, the network optimization center centrally manages the demand library for network construction and optimization, adopts a top-down management mode, and the demand types are mainly divided into coverage requirements and capacity requirements, including weak coverage complaints, measurement reports (MR) weak coverage, inferior to competitors, test black spots, key scenarios, simulated weak coverage, G network high traffic, G network high backflow, narrowband Internet of Things (NB-IoT), 5G and other requirements. The network optimization center distributes network optimization and network construction demand work orders from top to bottom based on the group's network performance standards.

[0003] When the demand is large, the provincial network optimization center is responsible for evaluating and formulating demand priorities, giving priority to meeting high-priority demands. Comprehensive value, complaints, MR, simulation and other dimensions are used to uniformly conduct rolling demand evaluation and scoring, distinguish the urgency of demand, and guide resource allocation. At the same time, the provincial network optimization center conducts rolling demand updates on a quarterly basis, uses big data intelligent analysis methods to output accurate demand, and improves demand quality.

[0004] Considering the future sixth generation mobile communication (6G) era, 6G networks will carry the diverse needs of thousands of industries. The network will provide traditional services for personal (2C) services, and will also provide diversified and customized services for various enterprise (2B) industries. 6G networks will have multiple key capabilities such as perception, communication, computing, and intelligence, and will have multiple heterogeneous resources such as communication, computing, and storage. Therefore, it is not enough to measure the network only with traditional CT domain performance indicators such as coverage and capacity, or to continue to perform operation and maintenance assessments for existing networks such as 2G to 5G according to existing ideas and performance indicators, while a set of performance indicators and demand generation and ranking models that can reflect network capabilities and service levels in multiple aspects are needed for 6G networks.

[0005] It can be seen that the existing technology for measuring network performance mainly focuses on coverage and capacity indicators, which can no longer meet the needs of multi-dimensional capability assessment of future 6G networks; in addition, the generation and sorting process of network optimization requirements still requires the intervention of expert experience and the dispatch of work orders, with high labor costs and no automation. Summary of the invention

[0006] At least one embodiment of the present invention provides a network optimization method, terminal and network equipment, which can realize automatic generation and sorting of optimization requirements and ensure network service quality.

[0007] According to one aspect of the present invention, at least one embodiment provides a network optimization method, including:

[0008] The network end-to-end management node establishes a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold;

[0009] According to the target mapping model, predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster;

[0010] The optimization demand cases of each optimization cluster are prioritized and sorted, and each optimization demand case is optimized in turn according to the sorting results.

[0011] In addition, according to at least one embodiment of the present invention, a target mapping model between node service performance data of each network service node under each optimization cluster and the total end-to-end service value is established, including:

[0012] Receive node service performance data of a network service node, the node service performance data including performance statistics of a molecular service and / or an atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service;

[0013] According to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract, a second mapping model between the end-to-end performance indicators of the end-to-end service and the end-to-end service value is established;

[0014] For each optimization cluster, the target mapping model of the node service performance data and the total end-to-end service value under the optimization cluster is established according to the first mapping model and the second mapping model.

[0015] In addition, according to at least one embodiment of the present invention, the present invention further comprises:

[0016] The correlation between the node service performance data of each network service node is analyzed to obtain the correlation coefficient of the molecular service performance between the network service nodes, and the network service nodes are divided into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient cutoff threshold.

[0017] In addition, according to at least one embodiment of the present invention, before training the first mapping model, the method further includes:

[0018] The network end-to-end management node sends a request message for end-to-end performance data of an end-to-end service to the service level performance evaluation module, and receives the end-to-end performance data of the end-to-end service sent by the service level performance evaluation module.

[0019] Furthermore, according to at least one embodiment of the present invention, the performance statistics of the molecular service of the network service node include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and a performance indicator associated with the characteristics of the molecular service, the performance indicator including at least one of accuracy, precision, and bit error rate;

[0020] The performance statistics of the atomic service of the network service node include at least one of the following: instantiation duration, number of instantiations, instantiation location, and load condition.

[0021] In addition, according to at least one embodiment of the present invention, the calculation and sorting of priorities of the optimization demand cases of each optimization cluster includes:

[0022] Based on at least one of the estimated value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized, and the recommended optimization completion time, the priority of each optimization demand case is calculated, and the optimization demand cases are sorted according to the priority.

[0023] In addition, according to at least one embodiment of the present invention, optimizing each optimization requirement case in sequence according to the sorting result includes:

[0024] According to the sorting result, a service performance improvement message is sent to the network service nodes in the optimization cluster, so that the network service nodes are optimized according to the service performance improvement message.

[0025] In addition, according to at least one embodiment of the present invention, the service performance improvement message includes at least one of the network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time.

[0026] According to another aspect of the present invention, at least one embodiment provides a network optimization method, including:

[0027] The network service node obtains the atomic service performance statistics from the atomic service management module, and measures and calculates the performance indicators of each molecular service to obtain molecular performance statistics;

[0028] The network service node sends the performance statistics of the molecular performance and / or atomic service to the network end-to-end management node;

[0029] The network service node receives a service performance improvement message sent by the network end-to-end management node, and performs performance optimization according to the service performance improvement message.

[0030] Furthermore, according to at least one embodiment of the present invention, the performance statistics of the molecular service include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and a performance indicator associated with the characteristics of the molecular service, the performance indicator including at least one of accuracy, precision, and bit error rate;

[0031] The performance statistics of the atomic service include at least one of the following: instantiation duration, number of instantiations, instantiation location, and load condition.

[0032] In addition, according to at least one embodiment of the present invention, the service performance improvement message includes at least one of a network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, a target value of the performance indicator to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time; the network service node performs performance optimization according to the service performance improvement message, including:

[0033] The network service node optimizes performance by executing at least one of the following configurations: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, and changing the scheduling priority of the molecular service.

[0034] In addition, according to at least one embodiment of the present invention, the network service node obtains the monitoring performance indicators corresponding to the molecular services, and measures and calculates the performance indicators of each molecular service based on the monitoring performance indicators, wherein the monitoring performance indicators include at least one of the following: a list of molecular service numbers, at least one performance indicator corresponding to each molecular service, a molecular service type, a molecular service subclass, a statistical calculation method for each performance indicator, and the importance of the performance indicator.

[0035] According to another aspect of the present invention, at least one embodiment provides a network end-to-end management node, characterized in that it includes:

[0036] A model building module, used to establish a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold;

[0037] A use case generation module, used to predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster according to the target mapping model, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate an optimization demand case for the optimization cluster;

[0038] The use case sorting module is used to calculate and sort the priority of the optimization demand cases of each optimization cluster, and optimize each optimization demand case in turn according to the sorting results.

[0039] In addition, according to at least one embodiment of the present invention, the model building module is specifically used to:

[0040] Receive node service performance data of a network service node, the node service performance data including performance statistics of a molecular service and / or an atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service;

[0041] According to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract, a second mapping model between the end-to-end performance indicators of the end-to-end service and the end-to-end service value is established;

[0042] For each optimization cluster, the target mapping model of the node service performance data and the total end-to-end service value under the optimization cluster is established according to the first mapping model and the second mapping model.

[0043] In addition, according to at least one embodiment of the present invention, the present invention further comprises:

[0044] The optimization cluster division module is used to analyze the correlation between the node service performance data of each network service node, obtain the correlation coefficient of the molecular service performance between the network service nodes, and divide the network service nodes into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient truncation threshold.

[0045] In addition, according to at least one embodiment of the present invention, the use case generation module is also used to send a request message for end-to-end performance data of the end-to-end service to the service level performance evaluation module before training the first mapping model, and receive the end-to-end performance data of the end-to-end service sent by the service level performance evaluation module.

[0046] In addition, according to at least one embodiment of the present invention, the use case sorting module is also used to calculate the priority of each optimization demand case based on at least one of the estimated value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized and the recommended optimization completion time, and sort the optimization demand cases according to the priority.

[0047] In addition, according to at least one embodiment of the present invention, the use case sorting module is further used to send a service performance improvement message to the network service node in the optimization cluster according to the sorting result, so that the network service node is optimized according to the service performance improvement message.

[0048] According to another aspect of the present invention, at least one embodiment provides a network end-to-end management node, including a transceiver and a processor, wherein:

[0049] The transceiver is used to establish a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold; predict the temporal changes of the total end-to-end service value or the temporal changes of the end-to-end performance index of each optimization cluster according to the target mapping model, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster; calculate and sort the priority of the optimization demand cases of each optimization cluster, and optimize each optimization demand case in turn according to the sorting result.

[0050] According to another aspect of the present invention, at least one embodiment provides a network end-to-end management node, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program implements the steps of the method described above when executed by the processor.

[0051] According to another aspect of the present invention, at least one embodiment provides a network service node, including:

[0052] The lifecycle management module is used to obtain the atomic service performance statistics from the atomic service management module, and measure and calculate the performance indicators of each molecular service to obtain molecular performance statistics;

[0053] A performance monitoring module, configured to send performance statistics of the molecular performance and / or atomic services to a network end-to-end management node;

[0054] The resource management module is used to receive the service performance improvement message sent by the network end-to-end management node, and perform performance optimization according to the service performance improvement message.

[0055] In addition, according to at least one embodiment of the present invention, the service performance improvement message includes at least one of a network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, a target value of the performance indicator to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time;

[0056] The resource management module is further used to optimize performance by executing at least one of the following configurations: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, and changing the scheduling priority of the molecular service.

[0057] In addition, according to at least one embodiment of the present invention, the performance monitoring module is also used to obtain monitoring performance indicators corresponding to molecular services, and measure and calculate the performance indicators of each molecular service based on the monitoring performance indicators, wherein the monitoring performance indicators include at least one of the following: a list of molecular service numbers, at least one performance indicator corresponding to each molecular service, a molecular service type, a molecular service subclass, a statistical calculation method for each performance indicator, and the importance of the performance indicator.

[0058] According to another aspect of the present invention, at least one embodiment provides a network service node, including a transceiver and a processor, wherein:

[0059] The transceiver is used to obtain atomic service performance statistics from the atomic service management module, and measure and calculate the performance indicators of each molecular service to obtain molecular performance statistics; and send the molecular performance and / or atomic service performance statistics to the network end-to-end management node;

[0060] The processor is used to receive a service performance improvement message sent by the network end-to-end management node, and perform performance optimization according to the service performance improvement message.

[0061] According to another aspect of the present invention, at least one embodiment provides a network service node, characterized in that it includes: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program implements the steps of the method described above when executed by the processor.

[0062] According to another aspect of the present invention, at least one embodiment provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the steps of the method described above are implemented.

[0063] Compared with the prior art, the network optimization method and device provided by the embodiments of the present invention can evaluate and rank the value of network optimization requirements, realize real-time monitoring of multi-dimensional network performance, and ensure continuous compliance with service quality standards through automatic generation and ranking of optimization requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0065] Figure 1 A schematic diagram of an application scenario of an embodiment of the present invention;

[0066] Figure 2 A flow chart of the network optimization method according to an embodiment of the present invention applied to a network end-to-end management node;

[0067] Figure 3 Another flow chart of the network optimization method according to the embodiment of the present invention applied to a network end-to-end management node;

[0068] Figure 4 An example diagram of establishing a target mapping model provided by an embodiment of the present invention;

[0069] Figure 5 A flow chart of a network optimization method according to an embodiment of the present invention applied to a network service node;

[0070] Figure 6 This is an example diagram of the network optimization method according to an embodiment of the present invention being applied to network slicing optimization;

[0071] Figure 7 A schematic diagram of the structure of a network end-to-end management node provided by an embodiment of the present invention;

[0072] Figure 8 Another structural diagram of a network end-to-end management node provided by an embodiment of the present invention;

[0073] Fig. 9 A schematic diagram of the structure of a network service node provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0074] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0075] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable in appropriate circumstances, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, the process, method, system, product or equipment comprising a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment. "And / or" in the specification and claims represents at least one of the connected objects.

[0076] The following description provides examples and does not limit the scope, applicability, or configuration set forth in the claims. Changes may be made to the functions and arrangements of the elements discussed without departing from the spirit and scope of the present disclosure. Various examples may appropriately omit, replace, or add various procedures or components. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.

[0077] As described in the background technology, the existing technology for measuring network performance mainly focuses on coverage and capacity indicators, which can no longer meet the needs of future multi-dimensional network capability assessment; in addition, the labor cost of generating and sorting network optimization requirements is high and automation has not been achieved.

[0078] To solve at least one of the above problems, the embodiment of the present invention provides a network performance index system model with the performance of molecular services as the main assessment dimension, and proposes a network optimization method that can predict and diagnose network performance degradation and evaluate and rank the value of network optimization requirements. Through the linkage of the above methods, real-time monitoring of the multi-dimensional performance of future 6G networks can be achieved, and the automatic generation and ranking of optimization requirements can be achieved to ensure that the service quality continues to meet the standards.

[0079] like Figure 1As shown, the network optimization method of the embodiment of the present invention is applied to a logical structure including a network end-to-end management node 12 (such as a central cloud) and a distributed network service node 11 (such as a wireless base station, a core network element). Figure 1 Only one network service node 11 is shown in the figure. It should be noted that there may be multiple network service nodes. Figure 1 The network service node 11 includes a molecular service management module, a molecular service generation module, an atomic service management module and a resource management module; the network end-to-end management node 12 includes a use case generation module and a use case sorting module. For the sake of simplicity, the network service node is sometimes referred to as a service node, and the network end-to-end management node is referred to as a management node.

[0080] Among them, the network service node 11 provides a calling interface for the end-to-end service through various atomic granularity services (atomic services). Considering that the network service node 11 has multiple capabilities (such as measurement, positioning, data forwarding, calculation, data collection, intelligent generation, etc.), the workflow of the end-to-end service can be disassembled into multiple molecular services (these molecular services can be executed in parallel or serially) according to the node capabilities, and executed inside multiple network service nodes. Each molecular service is instantiated as a combination of at least one atomic service and configured with relevant parameters. The relevant molecular service management module and molecular service generation module are also provided in the network service node.

[0081] The network end-to-end management node 12 is used for performance evaluation of end-to-end services, SLA management, network optimization demand generation, and network optimization queue maintenance. The network optimization method of the embodiment of the present invention will at least involve Figure 1 The molecular service management module (Molecular Service Management), use case generation module (Use case generator) and use case sorting module (Use case Queue) in it.

[0082] In the process of formulating network services, it is necessary to define the overall function set of the service, the smallest granularity of atomic service cutting, and the operation logic of each atomic service. For example, the overall function set in the base station is divided into multiple atomic services (such as applying for computing resources, executing computing tasks, applying for storage resources, calling intelligent models, training intelligent models, evaluating model performance, performing measurements, etc.), and a larger granularity service (referred to as molecular services in this article) can be provided by combining some atomic services. That is, atomic services are services of the first level of granularity, and molecular services are services of the second level of granularity, and the second level of granularity is greater than the first level of granularity; molecular services can be implemented by at least one atomic service, and a molecular service can be split into at least one atomic service.

[0083] Please refer to Figure 2 A network optimization method provided by an embodiment of the present invention, when applied to a network end-to-end management node, includes:

[0084] Step 100, the network end-to-end management node establishes a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold;

[0085] Step 200: predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster according to the target mapping model, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster;

[0086] Step 300 , calculating and sorting the priorities of the optimization demand cases of each optimization cluster, and optimizing each optimization demand case in turn according to the sorting results.

[0087] Through the above steps, the present invention can evaluate and sort the value of network optimization requirements, realize real-time monitoring of multi-dimensional network performance, and ensure service quality through automatic generation and sorting of optimization requirements.

[0088] Please refer to Figure 3 Another network optimization method provided by an embodiment of the present invention, when applied to a network end-to-end management node, includes:

[0089] Step 21, the network end-to-end management node receives the node service performance data of the network service node, wherein the node service performance data includes performance statistics of the molecular service and / or atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, trains and obtains a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service.

[0090] Here, the performance statistics of the molecular service of the network service node specifically include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and a performance indicator associated with the characteristics of the molecular service, wherein the performance indicator includes at least one of accuracy, precision, and bit error rate. The performance statistics of the atomic service of the network service node specifically include at least one of the following: instantiation time, number of instantiations, instantiation location, and load condition.

[0091] Before training the first mapping model, the management node may send a request message to the service level performance evaluation module, such as Figure 1The management node sends an SLP measurements request message to request the end-to-end performance data of the end-to-end service. The content of the message includes the end-to-end network service ID, or the network service name, or the network service group ID to be queried, and may also include auxiliary information related to the network service, the SLP performance index of the network service to be queried, and may also include the query time period, etc. Then, the management node receives the SLP measurements response message returned by the service level performance evaluation module to obtain the end-to-end performance data of the end-to-end service. The content of the message includes the end-to-end network service ID, or the network service name, or the network service group ID, the SLP performance index of the network service, the statistical results of the service performance index, and may also include parameters such as the statistical time period and the statistical confidence.

[0092] Step 22, the network end-to-end management node establishes a second mapping model between the end-to-end performance indicators of the end-to-end services and the end-to-end service values ​​according to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract.

[0093] Here, the management node can obtain the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end (E2E) service, as well as the value loss of breach of contract from the service level agreement (SLA) management module, thereby establishing a mapping model between the SLA performance indicators and the total value of the end-to-end service.

[0094] Step 23, the network end-to-end management node analyzes the correlation between the node service performance data of each network service node, obtains the correlation coefficient of the molecular service performance between the network service nodes, and divides the network service nodes into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient cutoff threshold; and for each optimization cluster, according to the first mapping model and the second mapping model, establishes a target mapping model between the node service performance data under the cluster and the total end-to-end service value.

[0095] Here, according to the correlation coefficient of the molecular service performance between the network service nodes, when the correlation of the molecular service performance of two service nodes exceeds a preset threshold, the two service nodes are added to the same optimization cluster, so that each service node is divided into a corresponding optimization cluster, and each optimization cluster includes at least one service node. Then, for each optimization cluster, according to the first mapping model and the second mapping model obtained in step 21 and step 22, a target mapping model of the node service performance data under the cluster and the total end-to-end service value is established. Figure 4 An example of establishing a target mapping model between the node service performance data of network service nodes 1 to m under a certain optimization cluster and the total end-to-end service value is given.

[0096] Step 24, based on the target mapping model, predict the time series changes of the total end-to-end service value or the time series changes of the end-to-end performance indicators of each optimization cluster, determine the optimization cluster whose value loss or end-to-end performance indicator deterioration exceeds the corresponding threshold, and generate an optimization demand case for the optimization cluster.

[0097] Here, the embodiment of the present invention uses the target mapping model to perform reasoning, predict the time series change of the value or the time series change of the SLA indicator of each optimization cluster, and generate corresponding optimization demand cases for the optimization clusters whose value loss exceeds the threshold. The content of the optimization demand case includes a node number set, a value loss estimate, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, and may also include a list of E2E services to be optimized, a recommended optimization completion time, etc.

[0098] Step 25, calculating and sorting the priorities of the optimization demand cases of each optimization cluster, and optimizing each optimization demand case in turn according to the sorting results.

[0099] Here, the management node can calculate the priority of each optimization demand case according to at least one of the estimated value of value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized, and the recommended optimization completion time, and sort the optimization demand cases according to the priority. Then, according to the sorting result, a service performance improvement message is sent to the network service node in the optimization cluster, so that the network service node is optimized according to the service performance improvement message, wherein the service performance improvement message includes at least one of the network service node number, the list of molecular services to be optimized, the list of performance indicators to be optimized, the target value of the performance indicator to be optimized, the numbers of other network service nodes in the same cluster, and the recommended optimization completion time.

[0100] Through the above steps, the embodiment of the present invention can evaluate and sort the value of network optimization requirements, realize real-time monitoring of the multi-dimensional performance of future 6G networks, and ensure continuous compliance with service quality standards through automatic generation and sorting of optimization requirements.

[0101] The network optimization method of the embodiment of the present invention is described below from the network service node side.

[0102] Please refer to Figure 5 The network optimization method of the embodiment of the present invention, when applied to a network service node, includes:

[0103] Step 41: The network service node obtains atomic service performance statistics from the atomic service management module, and measures and calculates the performance indicators of each molecular service to obtain molecular performance statistics.

[0104] Here, the performance statistics of the molecular service include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and performance indicators associated with the characteristics of the molecular service, and the performance indicators include at least one of accuracy, precision, and bit error rate; the performance statistics of the atomic service include at least one of the following: instantiation time, number of instantiations, instantiation location, and load conditions.

[0105] The network service node may obtain the monitoring performance indicators corresponding to each molecular service, and measure and calculate the performance indicators of each molecular service according to the monitoring performance indicators, wherein the monitoring performance indicators include at least one of the following: a molecular service number list, at least one performance indicator corresponding to each molecular service, a molecular service type, a molecular service subclass, a statistical calculation method for each performance indicator, and the importance of the performance indicator. For example, the network service node may obtain the monitoring performance indicators corresponding to each molecular service from a radio resource management (RRM) entity. Of course, in the embodiment of the present invention, the monitoring performance indicators corresponding to each molecular service may also be input by an operation and maintenance personnel.

[0106] Step 42: The network service node sends the performance statistics of the molecular performance and / or atomic service to the network end-to-end management node.

[0107] Here, the service node sends the performance statistics of the molecular performance and / or atomic service to the management node, so that the management node can perform modeling and analysis based on the performance data and generate optimization demand cases.

[0108] Step 43: the network service node receives a service performance improvement message sent by the network end-to-end management node, and performs performance optimization according to the service performance improvement message.

[0109] Here, the service performance improvement message includes at least one of the network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time.

[0110] The network service node optimizes performance by executing at least one of the following configurations: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, changing the scheduling priority of the molecular service, etc.

[0111] Through the above steps, the embodiment of the present invention can realize the automatic generation and sorting of network optimization requirements, and ensure that the service quality continues to meet the standards.

[0112] The network optimization method of the embodiment of the present invention is described above. Figure 1, the above process is introduced in more detail through the interaction processes and interaction messages between various entities.

[0113] Figure 1 The network service node 11 includes an atomic service management module, a molecular service generation module, a molecular service management module and a resource management module, wherein the molecular service management module specifically includes a life cycle management (LCM) module and a performance monitor (PM) module. The LCM module is used to manage the entire life cycle of the molecular service, including the initiation of the molecular service, the configuration of the relevant parameters of each atomic service included in a molecular service, the configuration of the relevant parameters of the molecular service, the change and termination of the molecular service, etc. The PM module is used to monitor the performance of the molecular service.

[0114] In the embodiment of the present invention, monitoring performance indicators of different molecular services are defined. The definition can be configured through an external management module or a resource management module (such as an RRM entity), or can be defined and written by an operation and maintenance personnel. The relevant messages include:

[0115] Performance metrics configuration message, which is sent by the external management module and / or the resource management module to the PM module. Its content includes: a list of molecular service numbers, one or more performance indicators corresponding to each molecular service, and may also include molecular service types, molecular service subclasses, statistical calculation methods for each indicator, importance of performance indicators, etc.

[0116] The LCM module obtains atomic service performance statistics from the atomic service management module, including instantiation duration, number of instantiations, instantiation location, load status, etc. It can also include metrics, traces or logs, etc. The relevant messages are as follows:

[0117] Atom service performance data request message, which is sent by the LCM module to the atomic service management module, is used to request atomic service performance statistics. Its content includes the number of the relevant atomic service (atom service id), the type of atomic service performance indicator to be obtained, the performance indicator list, and may also include the calling relationship of these atomic services (atom service) (such as parallel or serial, etc.), the correspondence between atomic services and molecular services, etc.

[0118] Atom service performance data response message: This message is sent by the atomic service management module to the LCM module. Its content includes the ID of the relevant atomic service (atom serviceid), the type of atomic service performance indicator returned, the performance indicator list, performance indicator statistics, and may also include the correspondence between atomic services and molecular services.

[0119] The PM module measures and calculates the performance indicators of each molecular service to obtain molecular performance statistics and corresponding atomic performance statistics. Here, the performance indicators that need to be measured and calculated, in addition to the monitoring indicators defined in the above performance indicator configuration message, may also include common basic performance indicators of molecular services, such as the time to complete the molecular service, the amount of resources consumed, and performance indicators associated with the characteristics of the molecular service, such as accuracy, precision, bit error rate, etc., depending on the content written by the operation and maintenance personnel or a third party.

[0120] The PM module sends the obtained molecular performance statistics and the corresponding atomic performance statistics to the use case generator module in the network end-to-end management node through the service performance measurements message. The relevant messages are as follows:

[0121] The service performance measurements message includes the molecular service number, molecular service performance indicator statistics, the atomic composition of the molecule, the atomic service number, and the atomic service performance indicator statistics. The content can be the source data of the above performance data or the form of these data after integration and preprocessing (after matrix transformation).

[0122] The use case generator module in the network end-to-end management node receives the end-to-end service quality evaluation data (service level performance measurements) sent by the service level performance evaluation module, and receives the service performance measurements sent by each network service node. According to the correspondence between the end-to-end service and the relevant network service nodes, the first mapping model of the node service performance data (service performance) and the end-to-end performance data (E2E service level performance) of the end-to-end service on these nodes is trained. The messages involved in the above process include:

[0123] The service level agreement measurement request (SLP measurements request) message is sent by the use case generation module to the service level performance evaluation (SLP evaluation) module to request the end-to-end service performance indicator measurement results from the SLP evaluation module. The content of the message includes the end-to-end network service ID or network service name or network service group ID to be queried, and may also include network service-related auxiliary information, the SLP performance indicators of the network service to be queried, and may also include the query time period, etc.

[0124] The service level agreement measurement response (SLP measurements response) message is sent by the SLPevaluation module to the use case generation module. Its content includes the end-to-end network service ID or network service name or network service group ID, the SLP performance indicators of the network service, the statistical results of the service performance indicators, and may also include the statistical time period, statistical confidence, etc.

[0125] The use case generation module obtains the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end (E2E) service and the value loss of breach of contract from the service level agreement management (SLA management) module, and establishes a second mapping model between the SLA performance indicators and the total value of the E2E service. The messages involved in this process include:

[0126] The service level agreement content request (SLA contents request) message is sent by the use case generation module to the SLA management module. The content of the message includes the end-to-end network service ID or network service name or network service group ID to be queried, and may also include auxiliary information related to the network service.

[0127] The service level agreement content response (SLA contents response) message is sent by the SLA management module to the use case generation module. The content of the message includes the end-to-end network service ID or network service name or network service group ID, and the corresponding SLA content (including service level related performance indicators, service level list, economic settlement methods corresponding to different service levels, etc.).

[0128] The use case generation module analyzes the correlation of performance indicators between nodes based on the node service performance data (service performance) reported by different network service nodes, obtains the correlation coefficient of molecular service performance between nodes, and combines the nodes into optimized clusters according to the correlation coefficient and the coefficient truncation threshold.

[0129] The use case generation module establishes a mapping relationship model between the service performance and the total E2E service value under each optimization cluster.

[0130] The use case generation module predicts the trend of molecular service performance indicators, performs reasoning based on the above model, predicts the time series changes in value or SLA indicators of each optimization cluster, and generates optimization demand cases for optimization clusters whose value loss exceeds the threshold; the content of the optimization demand case includes a set of node numbers, a value loss estimate, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, and may also include a list of E2E services to be optimized, and recommended optimization completion time, etc.

[0131] The use case generation module sends the optimized demand cases to the use case sorting module (use case queue).

[0132] The use case ranking module receives optimization demand cases and calculates and ranks the priority of each optimization demand case (use case). The priority calculation takes into account the value loss valuation, and may also take into account the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of E2E services to be optimized, the recommended optimization completion time, etc.

[0133] The use case sorting module sends a service performance improvement message (service performance to improve) to the network service nodes in the optimization cluster in queue order. The content of the message includes the node number, the list of molecular services to be optimized, the list of performance indicators to be optimized, the target value of the performance indicators to be optimized, and may also include the numbers of other nodes in the same cluster, the recommended optimization completion time, etc.

[0134] The resource management module (Resource Management) in the network service node receives the service performance improvement message (service performance to improve) sent by the use case sorting module, and performs root cause analysis on the molecular service to be optimized and the performance index according to the message, and formulates the optimization means. The resource management module implements the optimization means, including changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, changing the scheduling priority of the molecular service, etc. For example, the molecular service generation module can be instructed to execute at least one of the following changes: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, and changing the scheduling priority of the molecular service by sending a performance requirement message (Performance requirements) to the molecular service generation module.

[0135] Combine the following Figure 6, taking network slicing optimization as an example, the above method is explained.

[0136] Figure 6 This is an implementation case of the network optimization method of an embodiment of the present invention in a wireless mobile communication network. The base station adopts a microservice architecture, and atomic services are equivalent to various microservices. Microservices are combined into molecular services such as measurement, positioning, calculation, AI model generation, and data forwarding as the basic capabilities of the base station. The core network nodes also divide the service levels into different levels according to their own functions.

[0137] Vertical industries purchase end-to-end network slices from telecom network operators and sign SLAs with operators. The SLA signing management is in the network slice management system. In a certain area (a base station cluster with correlated molecular performance, or an area divided by an operator), there may be multiple network slices, involving multiple core network nodes (1...n) and base station nodes (1...n). In each node, the molecular service management module (molecular service management) combines microservices into molecular services, and then the network service orchestration module (not within the scope of the present invention) orchestrates the molecular services in each node required for the end-to-end service, forming a serial and parallel calling sequence of molecular services.

[0138] Within each node on this end-to-end path, molecular service management manages molecular-level services, collects performance data of molecular-level services, and collects performance data of related atomic services from the Kubernetes management entity. After processing, it is sent to the network slice service management system in the centralized network management and operation system (through the service Performance Measurements message).

[0139] The network slice service management system establishes a mapping model between the service performance data of each node in this area and the SLA indicators, as well as a mapping model between the SLA indicators and the total economic benefits of the network slice services in this area, and associates the two into a unified model. The network slice service management system can obtain performance measurement data directly from each node. If the amount of measurement data is large, it can also be compressed and transmitted, or distributed model training can be performed through distributed AI training. Therefore, the measurement data content transmitted in the service performance measurements message may be original measurement data or processed (after matrix transformation). The network slice service management system uses the model to predict the total economic benefits of all network slice services in this area, generate network optimization cases, and sort them. For each network optimization case, the molecular services and related performance indicators that need to be optimized on certain nodes are indicated to the nodes (through the serviceKPIs to be improved message).

[0140] In this example, the vertical industry service management system can configure the performance indicators, definitions, and calculation methods that need to be monitored by molecular services within the base station or core network node through the external interface, evaluate the end-to-end service performance of the network slice, and provide the evaluation results to the operator's network slice service management system.

[0141] From the above description, it can be seen that the network optimization method of the embodiment of the present invention can meet the needs of multi-dimensional network capability assessment, and can also realize the automatic generation and sorting of network optimization needs, thereby ensuring that the service quality continues to meet the standards.

[0142] The above describes various methods of the embodiments of the present invention. The following further provides devices for implementing the above methods.

[0143] Please refer to Figure 7 , the network end-to-end management node provided by the embodiment of the present invention includes:

[0144] A model building module 501 is used to build a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold;

[0145] A use case generation module 502 is used to predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster according to the target mapping model, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate an optimization demand case for the optimization cluster;

[0146] The use case sorting module 503 is used to calculate and sort the priority of the optimization demand cases of each optimization cluster, and optimize each optimization demand case in turn according to the sorting result.

[0147] Through the above modules, the present invention can evaluate and sort the value of network optimization requirements, realize real-time monitoring of multi-dimensional network performance, and ensure service quality through automatic generation and sorting of optimization requirements.

[0148] The model building module is specifically used to: receive node service performance data of a network service node, wherein the node service performance data includes performance statistics of the molecular service and / or atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model to obtain the node service performance data of each network service node and the end-to-end performance data of the end-to-end service; establish a second mapping model between the end-to-end performance indicator of the end-to-end service and the end-to-end service value according to the contract requirements of the end-to-end performance indicator corresponding to the service level of each end-to-end service and the value loss of breach of contract; for each optimization cluster, establish the target mapping model between the node service performance data under the optimization cluster and the total end-to-end service value according to the first mapping model and the second mapping model.

[0149] Optionally, the network end-to-end management node further includes:

[0150] The optimization cluster division module is used to analyze the correlation between the node service performance data of each network service node, obtain the correlation coefficient of the molecular service performance between the network service nodes, and divide the network service nodes into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient truncation threshold.

[0151] Optionally, the use case generation module is also used to send a request message for end-to-end performance data of the end-to-end service to the service level performance evaluation module before training the first mapping model, and receive the end-to-end performance data of the end-to-end service sent by the service level performance evaluation module.

[0152] Optionally, the performance statistics of the molecular service of the network service node include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and a performance indicator associated with the characteristics of the molecular service, wherein the performance indicator includes at least one of accuracy, precision, and bit error rate;

[0153] The performance statistics of the atomic service of the network service node include at least one of the following: instantiation duration, number of instantiations, instantiation location, and load condition.

[0154] Optionally, the use case sorting module is also used to calculate the priority of each optimization demand case based on at least one of the estimated value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized and the recommended optimization completion time, and sort the optimization demand cases according to the priority.

[0155] Optionally, the use case sorting module is further used to send a service performance improvement message to the network service nodes in the optimization cluster according to the sorting result, so that the network service nodes are optimized according to the service performance improvement message.

[0156] Optionally, the service performance improvement message includes at least one of the network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time.

[0157] It should be noted that the device in this embodiment is the same as the above Figure 2 The device corresponding to the method shown, the implementation methods in the above embodiments are all applicable to the embodiments of the device, and can also achieve the same technical effects. The above device provided by the embodiment of the present invention can implement all the method steps implemented by the above method embodiment, and can achieve the same technical effects. The parts and beneficial effects that are the same as those in the method embodiment will not be described in detail here.

[0158] Please refer to Figure 8 The embodiment of the present invention provides a schematic diagram of a structure of a network end-to-end management node 600, including: a processor 601, a transceiver 602, a memory 603 and a bus interface, wherein:

[0159] In the embodiment of the present invention, the network end-to-end management node 600 further includes: a program stored in the memory 603 and executable on the processor 601, and the program implements the following steps when executed by the processor 601:

[0160] Establishing a target mapping model between node service performance data of each network service node under each optimization cluster and total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold;

[0161] According to the target mapping model, predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster;

[0162] The optimization demand cases of each optimization cluster are prioritized and sorted, and each optimization demand case is optimized in turn according to the sorting results.

[0163] Optionally, when the processor executes the program, the processor further implements the following steps:

[0164] Receive node service performance data of a network service node, the node service performance data including performance statistics of a molecular service and / or an atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service;

[0165] According to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract, a second mapping model between the end-to-end performance indicators of the end-to-end service and the end-to-end service value is established;

[0166] For each optimization cluster, the target mapping model of the node service performance data and the total end-to-end service value under the optimization cluster is established according to the first mapping model and the second mapping model.

[0167] Optionally, when the processor executes the program, the processor further implements the following steps:

[0168] The correlation between the node service performance data of each network service node is analyzed to obtain the correlation coefficient of the molecular service performance between the network service nodes, and the network service nodes are divided into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient cutoff threshold.

[0169] Optionally, when the processor executes the program, the processor further implements the following steps:

[0170] Before training the first mapping model, a request message for end-to-end performance data of an end-to-end service is sent to a service level performance evaluation module, and the end-to-end performance data of the end-to-end service sent by the service level performance evaluation module is received.

[0171] Optionally, the performance statistics of the molecular service of the network service node include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and performance indicators associated with the characteristics of the molecular service, the performance indicators including at least one of accuracy, precision, and bit error rate; the performance statistics of the atomic service of the network service node include at least one of the following: instantiation duration, number of instantiations, instantiation location, and load conditions.

[0172] Optionally, when the processor executes the program, the processor further implements the following steps:

[0173] Based on at least one of the estimated value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized, and the recommended optimization completion time, the priority of each optimization demand case is calculated, and the optimization demand cases are sorted according to the priority.

[0174] Optionally, when the processor executes the program, the processor further implements the following steps:

[0175] According to the sorting result, a service performance improvement message is sent to the network service nodes in the optimization cluster, so that the network service nodes are optimized according to the service performance improvement message.

[0176] Optionally, the service performance improvement message includes at least one of the network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time.

[0177] It is understandable that in the embodiment of the present invention, when the computer program is executed by the processor 601, the above Figure 2 The various processes of the method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0178] exist Figure 8 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 601 and memory represented by memory 603. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 602 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium.

[0179] The processor 601 is responsible for managing the bus architecture and general processing, and the memory 603 can store data used by the processor 601 when performing operations.

[0180] It should be noted that the terminal in this embodiment is the same as the above Figure 2The device corresponding to the method shown, the implementation methods in the above embodiments are all applicable to the embodiments of the terminal, and can also achieve the same technical effect. In the device, the transceiver 602 and the memory 603, as well as the transceiver 602 and the processor 601 can be connected through the bus interface communication, the function of the processor 601 can also be implemented by the transceiver 602, and the function of the transceiver 602 can also be implemented by the processor 601. It should be noted that the above device provided by the embodiment of the present invention can implement all the method steps implemented by the above method embodiment, and can achieve the same technical effect, and the parts and beneficial effects that are the same as the method embodiment in this embodiment will not be specifically repeated here.

[0181] In some embodiments of the present invention, a computer-readable storage medium is further provided, on which a program is stored, and when the program is executed by a processor, the following steps are implemented:

[0182] Establishing a target mapping model between node service performance data of each network service node under each optimization cluster and total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold;

[0183] According to the target mapping model, predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster;

[0184] The optimization demand cases of each optimization cluster are prioritized and sorted, and each optimization demand case is optimized in turn according to the sorting results.

[0185] When the program is executed by the processor, it can implement all the implementation methods of the above-mentioned network optimization method applied to the network end-to-end management node, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0186] Please refer to Figure 1 The network service node provided by the embodiment of the present invention includes:

[0187] The lifecycle management module is used to obtain the atomic service performance statistics from the atomic service management module, and measure and calculate the performance indicators of each molecular service to obtain molecular performance statistics;

[0188] A performance monitoring module, configured to send performance statistics of the molecular performance and / or atomic services to a network end-to-end management node;

[0189] The resource management module is used to receive the service performance improvement message sent by the network end-to-end management node, and perform performance optimization according to the service performance improvement message.

[0190] Optionally, the performance statistics of the molecular service include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and a performance indicator associated with the characteristics of the molecular service, wherein the performance indicator includes at least one of accuracy, precision, and bit error rate;

[0191] The performance statistics of the atomic service include at least one of the following: instantiation duration, number of instantiations, instantiation location, and load condition.

[0192] Optionally, the service performance improvement message includes at least one of a network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, a target value of the performance indicator to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time; the resource management module is also used to perform performance optimization by executing at least one of the following configurations: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, and changing the scheduling priority of the molecular service.

[0193] Optionally, the performance monitoring module is also used to obtain monitoring performance indicators corresponding to molecular services, and measure and calculate the performance indicators of each molecular service based on the monitoring performance indicators, wherein the monitoring performance indicators include at least one of the following: a list of molecular service numbers, at least one performance indicator corresponding to each molecular service, a molecular service type, a molecular service subclass, a statistical calculation method for each performance indicator, and the importance of the performance indicator.

[0194] It should be noted that the device in this embodiment is the same as the above Figure 5 The device corresponding to the method shown, the implementation methods in the above embodiments are all applicable to the embodiments of the device, and can also achieve the same technical effects. It should be noted that the above device provided in the embodiment of the present invention can implement all the method steps implemented in the above method embodiment, and can achieve the same technical effects, and the parts and beneficial effects that are the same as those in the method embodiment in this embodiment will not be specifically described here.

[0195] Please refer to Fig. 9 The embodiment of the present invention provides a schematic diagram of a structure of a network service node 700, including: a processor 701, a transceiver 702, a memory 703 and a bus interface, wherein:

[0196] In the embodiment of the present invention, the network service node 700 further includes: a program stored in the memory 703 and executable on the processor 701, and the program implements the following steps when executed by the processor 701:

[0197] Obtaining atomic service performance statistics from the atomic service management module, and measuring and calculating the performance indicators of each molecular service to obtain molecular performance statistics;

[0198] Sending performance statistics of the molecular performance and / or atomic service to a network end-to-end management node;

[0199] Receive a service performance improvement message sent by the network end-to-end management node, and perform performance optimization according to the service performance improvement message.

[0200] Optionally, the performance statistics of the molecular service include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and performance indicators associated with the characteristics of the molecular service, the performance indicators including at least one of accuracy, precision, and bit error rate; the performance statistics of the atomic service include at least one of the following: instantiation time, number of instantiations, instantiation location, and load conditions.

[0201] Optionally, the service performance improvement message includes at least one of a network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, a target value of the performance indicator to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time; when the processor executes the program, the following steps are also implemented: performance optimization is performed by executing at least one of the following configurations: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, and changing the scheduling priority of the molecular service.

[0202] Optionally, when the processor executes the program, the processor further implements the following steps:

[0203] Acquire monitoring performance indicators corresponding to the molecular services, and measure and calculate the performance indicators of each molecular service according to the monitoring performance indicators, wherein the monitoring performance indicators include at least one of the following: a list of molecular service numbers, at least one performance indicator corresponding to each molecular service, a molecular service type, a molecular service subclass, a statistical calculation method for each performance indicator, and an importance of the performance indicator.

[0204] It is understandable that in the embodiment of the present invention, when the computer program is executed by the processor 701, the above Figure 5 The various processes of the method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0205] exist Fig. 9In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 701 and memory represented by memory 703. The bus architecture may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 702 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium.

[0206] The processor 701 is responsible for managing the bus architecture and general processing, and the memory 703 can store data used by the processor 701 when performing operations.

[0207] It should be noted that the terminal in this embodiment is the same as the above Figure 5 The device corresponding to the method shown, the implementation methods in the above embodiments are all applicable to the embodiments of the terminal, and can also achieve the same technical effect. In the device, the transceiver 702 and the memory 703, as well as the transceiver 702 and the processor 701 can be connected through the bus interface communication, the function of the processor 701 can also be implemented by the transceiver 702, and the function of the transceiver 702 can also be implemented by the processor 701. It should be noted that the above device provided by the embodiment of the present invention can implement all the method steps implemented by the above method embodiment, and can achieve the same technical effect, and the parts and beneficial effects that are the same as the method embodiment in this embodiment will not be specifically repeated here.

[0208] In some embodiments of the present invention, a computer-readable storage medium is further provided, on which a program is stored, and when the program is executed by a processor, the following steps are implemented:

[0209] Obtaining atomic service performance statistics from the atomic service management module, and measuring and calculating the performance indicators of each molecular service to obtain molecular performance statistics;

[0210] Sending performance statistics of the molecular performance and / or atomic service to a network end-to-end management node;

[0211] Receive a service performance improvement message sent by the network end-to-end management node, and perform performance optimization according to the service performance improvement message.

[0212] When the program is executed by the processor, it can implement all the implementation methods of the above-mentioned network optimization method applied to the network service node, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0213] Those skilled 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 the present invention.

[0214] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0215] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. 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.

[0216] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0217] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0218] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0219] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A network optimization method, characterized in that: include: The network end-to-end management node establishes a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold; According to the target mapping model, predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster; Calculate and sort the priorities of the optimization demand cases of each optimization cluster, and optimize each optimization demand case in turn according to the sorting results; Among them, a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value is established, including: Receive node service performance data of a network service node, the node service performance data including performance statistics of a molecular service and / or an atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service; According to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract, a second mapping model between the end-to-end performance indicators of the end-to-end service and the end-to-end service value is established; For each optimization cluster, the target mapping model of the node service performance data and the total end-to-end service value under the optimization cluster is established according to the first mapping model and the second mapping model.

2. The method according to claim 1, characterized in that Also includes: The correlation between the node service performance data of each network service node is analyzed to obtain the correlation coefficient of the molecular service performance between the network service nodes, and the network service nodes are divided into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient cutoff threshold.

3. The method according to claim 1, characterized in that Before training the first mapping model, the method further includes: The network end-to-end management node sends a request message for end-to-end performance data of an end-to-end service to the service level performance evaluation module, and receives the end-to-end performance data of the end-to-end service sent by the service level performance evaluation module.

4. The method according to claim 1, characterized in that The performance statistics of the molecular service of the network service node include at least one of the following: the time to complete the molecular service, the amount of resources consumed, and a performance indicator associated with the characteristics of the molecular service, wherein the performance indicator includes at least one of accuracy, precision, and bit error rate; The performance statistics of the atomic service of the network service node include at least one of the following: instantiation duration, number of instantiations, instantiation location, and load condition.

5. The method according to claim 1, characterized in that The priority calculation and sorting of the optimization demand cases of each optimization cluster includes: Based on at least one of the estimated value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized, and the recommended optimization completion time, the priority of each optimization demand case is calculated, and the optimization demand cases are sorted according to the priority.

6. The method according to claim 1, characterized in that The optimization of each optimization requirement case is performed in turn according to the sorting results, including: According to the sorting result, a service performance improvement message is sent to the network service nodes in the optimization cluster, so that the network service nodes are optimized according to the service performance improvement message.

7. The method according to claim 6, characterized in that The service performance improvement message includes at least one of a network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time.

8. The method according to claim 6, characterized in that Also includes: The network service node receives a service performance improvement message sent by the network end-to-end management node, and performs performance optimization according to the service performance improvement message.

9. The method according to claim 8, characterized in that The service performance improvement message includes at least one of the network service node number, a list of molecular services to be optimized, a list of performance indicators to be optimized, target values ​​of performance indicators to be optimized, numbers of other network service nodes in the same cluster, and a recommended optimization completion time; The network service node performs performance optimization according to the service performance improvement message, including: The network service node optimizes performance by executing at least one of the following configurations: changing the structural configuration of the molecular service, changing the underlying resources called by the molecular service, changing the location of the molecular service instantiation, and changing the scheduling priority of the molecular service.

10. The method according to claim 1, characterized in that Also includes: The network service node obtains monitoring performance indicators corresponding to the molecular services, and measures and calculates performance indicators of each molecular service according to the monitoring performance indicators, wherein the monitoring performance indicators include at least one of the following: a molecular service number list, at least one performance indicator corresponding to each molecular service, a molecular service type, a molecular service subclass, a statistical calculation method for each performance indicator, and an importance of the performance indicator.

11. A network end-to-end management node, characterized in that: include: A model building module, used to establish a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold; A use case generation module, used to predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster according to the target mapping model, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate an optimization demand case for the optimization cluster; The use case sorting module is used to calculate and sort the priority of the optimization demand cases of each optimization cluster, and optimize each optimization demand case in turn according to the sorting results; Wherein, the model building module is specifically used for: Receive node service performance data of a network service node, the node service performance data including performance statistics of a molecular service and / or an atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service; According to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract, a second mapping model between the end-to-end performance indicators of the end-to-end service and the end-to-end service value is established; For each optimization cluster, the target mapping model of the node service performance data and the total end-to-end service value under the optimization cluster is established according to the first mapping model and the second mapping model.

12. The network end-to-end management node according to claim 11, characterized in that: Also includes: The optimization cluster division module is used to analyze the correlation between the node service performance data of each network service node, obtain the correlation coefficient of the molecular service performance between the network service nodes, and divide the network service nodes into at least one optimization cluster according to the correlation coefficient and the corresponding coefficient truncation threshold.

13. The network end-to-end management node according to claim 11, characterized in that: The use case generation module is also used to send a request message for end-to-end performance data of the end-to-end service to the service level performance evaluation module before training the first mapping model, and receive the end-to-end performance data of the end-to-end service sent by the service level performance evaluation module.

14. The network end-to-end management node according to claim 11, characterized in that: The use case sorting module is also used to calculate the priority of each optimization demand case based on at least one of the estimated value loss, the number of nodes to be optimized, the list of molecular services to be optimized, the list of performance indicators to be optimized, the list of end-to-end services to be optimized and the recommended optimization completion time, and sort the optimization demand cases according to the priority.

15. The network end-to-end management node according to claim 11, characterized in that: The use case sorting module is further used to send a service performance improvement message to the network service node in the optimization cluster according to the sorting result, so that the network service node is optimized according to the service performance improvement message.

16. A network end-to-end management node, characterized in that: comprising a transceiver and a processor, wherein: The processor is used to establish a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value, wherein each optimization cluster includes a plurality of network service nodes whose correlation coefficients of molecular service performances exceed a preset threshold; predict the time series change of the total end-to-end service value or the time series change of the end-to-end performance index of each optimization cluster according to the target mapping model, determine the optimization cluster whose value loss or end-to-end performance index deterioration exceeds the corresponding threshold, and generate the optimization demand case of the optimization cluster; calculate and sort the priority of the optimization demand cases of each optimization cluster, and optimize each optimization demand case in turn according to the sorting result; Among them, a target mapping model between the node service performance data of each network service node under each optimization cluster and the total end-to-end service value is established, including: Receive node service performance data of a network service node, the node service performance data including performance statistics of a molecular service and / or an atomic service of the network service node; and, according to the correspondence between the end-to-end service and each network service node, train a first mapping model between the node service performance data of each network service node and the end-to-end performance data of the end-to-end service; According to the contract requirements of the end-to-end performance indicators corresponding to the service levels of each end-to-end service and the value loss of breach of contract, a second mapping model between the end-to-end performance indicators of the end-to-end service and the end-to-end service value is established; For each optimization cluster, the target mapping model of the node service performance data and the total end-to-end service value under the optimization cluster is established according to the first mapping model and the second mapping model.

17. A network end-to-end management node, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

18. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images