Method, device and storage medium for realizing user intention in intention-driven network

Abstract

The present disclosure proposes a method, device and storage medium for realizing user intent in an intent-driven network, and relates to the field of network technology. The present disclosure provides a method for realizing user intent in an intent-driven network, including: obtaining a user intent set; classifying user intent in the user intent set to obtain classified user intent; converting the classified user intent into a network policy; verifying the network policy by executing the network policy in a network environment to obtain a verification result; confirming or modifying the network policy according to the verification result until the user intent is achieved. Through such a method, before converting the intent into a network policy, the intent is first classified, thereby reducing the amount of data processed by retrieval and polling in the policy library, reducing time complexity, and improving processing efficiency; reducing the probability of erroneous conversion and improving accuracy.

Description

Technical Field

[0001] The present disclosure relates to the field of network technology, and in particular to a method, device and storage medium for realizing user intent in an intent-driven network. Background Art

[0002] With the vigorous development of the Internet and applications, the network is transforming towards a more intelligent direction. How to recognize the intention of user input methods such as text, voice, GUI (Graphical User Interface), and third-party input, convert it into an abstract network language, and issue configuration and execute it in the system is the inevitable development trend of future intelligent networks. The network based on intent recognition can be summarized as a closed-loop network architecture that automatically builds and operates based on human business intentions while mastering the global state of the network. Summary of the invention

[0003] One object of the present disclosure is to improve the efficiency and accuracy of intention realization.

[0004] According to one aspect of some embodiments of the present disclosure, a method for realizing user intent in an intent-driven network is proposed, including: obtaining a user intent set; classifying the user intent in the user intent set to obtain the classified user intent; converting the classified user intent into a network policy; verifying the network policy by executing the network policy in a network environment to obtain a verification result; confirming or modifying the network policy according to the verification result until the user intent is achieved.

[0005] In some embodiments, user intent is classified, and obtaining classified user intent includes: determining the business scenario, user type, and intent type of each user intent in the user intent set, wherein the intent type includes intents with a life cycle and intents without a life cycle.

[0006] In some embodiments, determining the business scenario of each user intent in the user intent set includes: obtaining the identifier of the user who provides the user intent; determining the network environment in which the user is located based on the user identifier, and taking the network environment as the business scenario of the user intent, wherein the network environment includes an operator network, a DC (Data Center) network, or an enterprise network.

[0007] In some embodiments, determining the user type of each user intent in a user intent set includes: extracting key fields of the user intent through an attribute collector; a function judger determines whether the user intent is feedback-oriented based on the key fields; if the user intent is not feedback-oriented, determining the user type as an ordinary user; if the user intent is feedback-oriented, determining the user type as an administrator.

[0008] In some embodiments, determining the intent type of each user intent in a user intent set includes: extracting key fields of the user intent through an attribute collector; a function judger determines whether the user intent has a life cycle based on the key fields; if the user intent has a life cycle, determining the intent type as an intent with a life cycle; if the user intent does not have a life cycle, determining the intent type as an intent without a life cycle.

[0009] In some embodiments, converting the classified user intent into a network policy includes: determining a corresponding policy library according to the classified user intent; and converting the classified user intent into a network policy based on the determined policy library.

[0010] In some embodiments, the network policy is verified by executing the network policy in the network environment, and obtaining the verification result includes: executing the network policy in the network environment; generating an executable verification result based on whether the network policy is successfully executed; generating a validity verification result based on whether the network status change meets the user's expected intention; wherein the verification result includes an executable verification result and a validity verification result.

[0011] In some embodiments, confirming or modifying the network policy according to the verification results until the user intent is achieved includes: based on the verification results, based on artificial intelligence or administrator instructions, screening out network policies that fail the verification; determining the classified user intent corresponding to the network policies that fail the verification, and re-executing the operation of converting the classified user intent into the network policy until the verification passes.

[0012] In some embodiments, the method for implementing user intent in an intent-driven network further includes: after obtaining a user intent set, verifying the user intent in the user intent set; feeding back the user intent that fails the verification to the user so that the user can modify it; and performing an operation of classifying the user intent set on the user intent that passes the verification.

[0013] Through this method, before converting the intent into a network policy, the intent is first classified, thereby reducing the amount of data processed by retrieval and polling in the policy library, reducing time complexity, and improving processing efficiency; reducing the probability of erroneous conversion and improving accuracy.

[0014] According to one aspect of some embodiments of the present disclosure, a device for realizing user intent in an intent-driven network is proposed, comprising: an intent acquisition unit, configured to acquire a set of user intents; a classification unit, configured to classify the set of user intents and acquire classified user intents; a conversion unit, configured to convert the classified user intents into network policies; a verification unit, configured to verify a network policy by executing the network policy in a network environment and acquire a verification result; and a decision unit, configured to confirm or modify the network policy according to the verification result until the user intent is achieved.

[0015] In some embodiments, the method for implementing user intent in an intent-driven network also includes: an intent management unit, configured to verify the user intent in the user intent set obtained by the intent acquisition unit; feedback the user intent that fails the verification to the user so that the user can modify it; and send the user intent that passes the verification to the classification unit.

[0016] According to one aspect of some embodiments of the present disclosure, a device for realizing user intent in an intent-driven network is proposed, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute any one of the user intent realization methods in the intent-driven network mentioned above based on instructions stored in the memory.

[0017] Such a device can classify intents before converting them into network policies, thereby reducing the amount of data processed by retrieval and polling in the policy library, reducing time complexity, and improving processing efficiency; reducing the probability of erroneous conversion and improving accuracy.

[0018] According to one aspect of some embodiments of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored, and when the instructions are executed by a processor, the steps of any one of the above-mentioned methods for realizing user intent in an intent-driven network are implemented.

[0019] By executing instructions on such a storage medium, the intent can be classified before converting it into a network policy, thereby reducing the amount of data processed by retrieval and polling in the policy library, reducing time complexity, and improving processing efficiency; reducing the probability of erroneous conversion and improving accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0021] Figure 1 The present invention is a flowchart of some embodiments of the method for realizing user intent in an intent-driven network of the present invention.

[0022] Figure 2 The present invention is a flowchart of some embodiments of classifying user intent sets and converting user intent in the method for realizing user intent in an intent-driven network of the present invention.

[0023] Figure 3 The present invention provides flowcharts of other embodiments of the method for realizing user intent in an intent-driven network of the present invention.

[0024] Figure 4ASchematic diagram of some embodiments of an apparatus for realizing user intent in an intent-driven network of the present disclosure.

[0025] Figure 4B The present invention is a schematic diagram of some embodiments of the operating logic of a user intent implementation device in an intent-driven network of the present invention.

[0026] Figure 5 A schematic diagram of some embodiments of a classification unit in a device for implementing user intent in an intent-driven network of the present disclosure.

[0027] Figure 6 Schematic diagrams of other embodiments of the apparatus for realizing user intent in an intent-driven network of the present disclosure.

[0028] Figure 7 Schematic diagrams of some further embodiments of the apparatus for realizing user intent in an intent-driven network of the present disclosure. DETAILED DESCRIPTION

[0029] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments.

[0030] Intent-driven networks involve several key steps. The first is to be able to obtain the network requirements proposed by users, namely intents, translate the received intents into network policies, and verify the enforceability of the policies based on the current network status. The policies that have passed the verification will then be issued to the actual network. In addition, the system must monitor the network status in real time to ensure that the user's intent is correctly implemented and feedback the results to the user. Key technologies include intent analysis and translation, policy issuance and execution, and network status information collection and feedback, so that more flexible network control and policies can be easily loaded on the basic network. Therefore, it is of great significance to study the key technologies and implementations of correctly converting user intent into the correct network language and issuing configurations in intent-driven networks.

[0031] The inventors found that after the user inputs the intent, the intent network retrieves and polls a large amount of data in the policy library, extracts key information and converts it into an intent delivery strategy. The time complexity is high and the system operation efficiency is low. In addition, due to the lack of an effective intent classification mechanism, there is a high risk factor for how to correctly translate the intent into network language for execution, especially for more complex user intents, and the user experience cannot be guaranteed.

[0032] The flowcharts of some embodiments of the method for realizing user intent in an intent-driven network of the present disclosure are as follows: Figure 1 shown.

[0033] In step 101, a user intent set is obtained. In some embodiments, one or more user intents provided by a user are obtained through a business layer, and the user intent set includes all intents provided by the user.

[0034] In some embodiments, after obtaining the intents provided by the user, these intents can be verified and classified through the intent orchestration layer, and qualified intents are further processed in step 102, and unqualified intents are fed back to the user for further modification and adjustment.

[0035] In step 102, the user intentions in the user intention set are classified to obtain classified user intentions. In some embodiments, user intentions can be classified from three aspects: business scenario, user type, and business attribute.

[0036] The business scenario can be the current network environment, such as operator network, DC network or enterprise network; user types can be divided into managers or operation operators (collectively referred to as managers in this article) who need to obtain network status feedback, and ordinary users or developers who do not need to obtain network status feedback; business attributes can be divided into intents without a life cycle, in which the intent is completed after the specified operation is successfully executed, and intents with a life cycle, in which once the intent is successfully activated and deployed, the system will keep all related intents active until they are deactivated or deleted.

[0037] In step 103, the classified user intent is converted into a network policy. In some embodiments, each classification has a corresponding policy library, and the corresponding policy library can be determined according to the classified user intent, and the classified user intent is converted into a network policy based on the determined policy library. In some embodiments, different classifications have corresponding converters, and the corresponding converters can be determined according to the user intent to convert the classified user intent into a network policy, and different converters are based on different policy libraries.

[0038] In step 104, the network policy is verified by executing the network policy in the network environment to obtain a verification result. In some embodiments, the verification result is for each user policy.

[0039] In step 105, it is determined whether the network policy corresponding to the verification result passes according to the verification result. For the network policy that passes the verification, step 106 is executed; for the policy that fails the verification, it needs to be modified. In some embodiments, step 103 can be executed to reconvert the corresponding intent into a network policy.

[0040] In step 106, a network policy is determined.

[0041] Through this method, before converting the intent into a network policy, the intent is first classified, thereby reducing the amount of data processed by retrieval and polling in the policy library, reducing time complexity, and improving processing efficiency; reducing the probability of erroneous conversion and improving accuracy.

[0042] In some embodiments, the verification of network policies is divided into two types, one is the configuration executable verification. The configuration obtained in the translation process needs to be verified whether it can be executed in the actual network environment based on the real-time network status. The second is validity verification. When the configuration is actually executed, the network status will change. At this time, it is necessary to verify whether the execution of the configuration causes the network status to change as expected. In some embodiments, the process of generating verification results may include: executing the network policy in the network environment; generating an executable verification result based on whether the network policy is successfully executed; generating a validity verification result based on whether the change in network status meets the user's expectations, wherein the verification results include an executable verification result and a validity verification result.

[0043] Through this method, the generated strategy can be verified from two aspects: feasibility and effectiveness, so as to make a reasonable evaluation of the network strategy and provide a data basis for subsequent decision-making, which not only ensures that it can be executed, but also ensures that the execution results are in line with the user's intentions, thereby improving the accuracy of the intention realization.

[0044] In some embodiments, determining the business scenario of each user intent in the user intent set includes: obtaining the identifier of the user who provides the user intent; determining the network environment in which the user is located based on the user identifier, and taking the network environment as the business scenario of the user intent, wherein the network environment includes an operator network, a DC network, or an enterprise network.

[0045] In some embodiments, determining the user type of each user intent in a user intent set includes: extracting key fields of the user intent through an attribute collector; a function judger determines whether the user intent is feedback-oriented based on the key fields; if the user intent is not feedback-oriented, determining the user type as an ordinary user; if the user intent is feedback-oriented, determining the user type as an administrator.

[0046] In some embodiments, determining the intent type of each user intent in a user intent set includes: extracting key fields of the user intent through an attribute collector; a function judger determines whether the user intent has a life cycle based on the key fields; if the user intent has a life cycle, determining the intent type as an intent with a life cycle; if the user intent does not have a life cycle, determining the intent type as an intent without a life cycle.

[0047] In some embodiments, confirming or modifying the network policy according to the verification results until the user intent is achieved includes: based on the verification results, based on artificial intelligence or administrator instructions, screening out network policies that fail the verification; determining the classified user intent corresponding to the network policies that fail the verification, and re-executing the operation of converting the classified user intent into the network policy until the verification passes.

[0048] The flowchart of some embodiments of the user intent set classification and user intent conversion in the user intent realization method in the intent-driven network disclosed in the present invention is as follows: Figure 2 shown.

[0049] In step 201, a user identifier user-id is obtained.

[0050] In step 202, the service scenario is determined according to the user identifier. The basic information of the user can be identified according to the user-id, so as to determine the network environment from which the user comes, such as the operator network, DC network or enterprise network, as the service scenario.

[0051] In step 203, a service scenario identifier (intent-service identifier) ​​of the corresponding service scenario is added to the user intent message.

[0052] In step 204, it is determined whether the user's intention is feedback-oriented. In some embodiments, the key fields of the intention can be retrieved through the attribute collector to determine whether the intention is feedback-oriented.

[0053] Feedback refers to whether network information needs to be fed back to the proposer after the intent is executed. For ordinary users and application developers, they do not need to know the network details and underlying configuration information, and do not care about how the intent is executed. They only care about whether the execution result of the intent achieves the goal, the completion quality, the execution time, etc. Therefore, the user intent provided by these people does not have feedback. For managers, such as network administrators and operation operators, they execute intents, such as allocating network resources, selecting transmission paths, handling network failures, etc., and they need multiple feedback indicators for the network resource conditions, congestion conditions, and failure conditions after the intent is executed. Therefore, the user intent provided by these people has feedback.

[0054] If the user's intention is feedback-oriented, step 205 is executed; if the user's intention is not feedback-oriented, step 206 is executed.

[0055] In step 205, the user type is an operator or an administrator.

[0056] In step 206, the user type is a developer or a common user.

[0057] In step 207, a user type identifier (intent-user identifier) ​​of a corresponding type is added to the user intent message.

[0058] In step 208, it is determined whether each user intention has a life cycle. If the user intentions with a life cycle are screened out, step 210 is executed; if the user intentions without a life cycle are screened out, step 209 is executed.

[0059] The lifecycle is used to indicate whether the user intent is short-lived or persistent. For intents without a lifecycle, the intent is completed after the specified action is successfully performed and no longer affects the target object; for intents with a lifecycle, once the intent is successfully activated and deployed, the system will keep all related intents active until they are deactivated or deleted.

[0060] In step 209, the user intent has no life cycle and belongs to services such as configuration and resource request, and a service attribute identifier with no life cycle needs to be added.

[0061] In step 210, the user intent has a life cycle, and services with business attributes such as network optimization, connection, and security need to be added with an intent type identifier with a life cycle.

[0062] In step 211, a corresponding intent type identifier (intent-type identifier) ​​is added for each user intent.

[0063] In some embodiments, the classification logic may be as shown in Table 1 below:

[0064] In some embodiments, the relationship between the submodules in the classification unit can be as follows:

[0065] Table 1 Classification relationship examples

[0066]

[0067] In step 212, based on the added various identification header combinations, the intended business scenario, user type and intent type are clarified, so as to accurately retrieve the corresponding policy library and output it to the corresponding converter module for configuration and distribution.

[0068] Through this method, the generation of intent-based policies can be refined. By extracting key information about user intent, the intent classification device can manage corresponding classification rules and add identification. This allows for more accurate retrieval of the corresponding policy library, generation of configuration policies and distribution, and improved reliability of intent implementation.

[0069] In some embodiments, through intent classification, in addition to being able to more accurately search the intent policy library, more convenient and faster network management requests can also be provided for different users in different business scenarios, as shown in Table 2:

[0070] Table 2 Business scenario classification and user type

[0071]

[0072] Taking the cloud administrator as an example, the flowcharts of other embodiments of the method for realizing user intent in an intent-driven network of the present disclosure are as follows: Figure 3 shown.

[0073] In step 301 , a cloud administrator provides user intent.

[0074] In step 302, the user intents in the user intent set are verified; the user intents that fail the verification are fed back to the cloud administrator for modification; and step 303 is executed for the user intents that pass the verification.

[0075] In step 303, the user-id is obtained.

[0076] In step 304, it is determined that the service scenario is a DC network according to the user-id.

[0077] In step 305, the intent-service identifier corresponding to the DC network is added.

[0078] In step 306, it is determined whether the user's intention is feedback-oriented. If the determination result is that the user's intention is feedback-oriented, step 307 is executed.

[0079] In step 307, it is determined that the user type is an administrator.

[0080] In step 308, the intent-user identifier corresponding to the administrator is added.

[0081] In step 309, it is determined whether the user intention has a life cycle.

[0082] For the intent that the cloud administrator wants to execute, such as load balancing: for all flows that require NFV (Network Functions Virtualization) service links, it is desired to limit the maximum load of any VNF (Virtualised Network Function) node / container to less than 50%, and limit the maximum load of any network link to less than 70%. The attribute collector retrieves the load balancing keyword, determines that the intent is a network optimization intent requirement with a lifecycle, and executes step 311.

[0083] In addition, the cloud administrator may also provide some user intentions that do not have a life cycle and execute step 310 .

[0084] In step 310, the intent without life cycle is screened out, such as configuration, resource request and other services.

[0085] In step 311, the intentions with a life cycle are screened out, including network optimization, connection, security and other services.

[0086] In step 312, the intent-type identifier corresponding to the life cycle is added.

[0087] In step 313, a policy library and a converter corresponding to the intention conversion are determined, and the user intentions are converted into network policies respectively.

[0088] Through this method, the user's business scenario, user type and business attributes are gradually refined, and the corresponding policy library is searched according to the various identification headers of the intent, and converted to the corresponding policy converter for configuration and distribution. While improving the accuracy of intent translation, it also reduces the interaction between intent conversion and verification, thereby improving system efficiency and enhancing user experience.

[0089] In some embodiments, since different business scenarios and different users have different intent requirements, for cloud administrator users, intent types can be divided into the following modules:

[0090] 1. Cloud management intent: configuration of VMs (Virtual Machines), databases, application servers, and network connectivity requests, management of communications between VMs, etc.

[0091] 2. Resource management intention: Automatically allocate or optimize VM resources such as computing and storage based on cloud management system data analysis;

[0092] 3. Operational task intent: automated execution tasks other than basic configuration and resource management, such as site selection, transmission path, fault handling, network congestion, etc.;

[0093] 4. Policy intent: Automate and lifecycle management of administrators’ routine operational tasks, such as load balancing.

[0094] After retrieving the basic information of the intent, search the corresponding policy library based on the key fields. Different intent types have different attributes. Search the cloud administrator policy library through the key fields to effectively identify the intent type. The key fields include:

[0095] Intent attributes: C1 = Connectivity; C2 = Security; C3 = QoS; C4 = Compute resources; C5 = Storage resources

[0096] Network resources: C1 = virtual resources; C2 = physical resources

[0097] Abstractness: C1 = with feedback; C2 = without feedback

[0098] Life cycle: C1 = full life cycle; C2 = short-term intention

[0099] Table 3. Examples of key field identification for cloud management cloud users

[0100]

[0101] Based on this table, key fields are extracted and retrieved, thereby completing the strategy library for various business scenarios.

[0102] Schematic diagrams of some embodiments of the user intention realization device in the intention-driven network of the present disclosure are as follows Figure 4A shown.

[0103] The intention acquisition unit 401 can acquire a user intention set. In some embodiments, one or more user intentions provided by the user are acquired through the business layer, and the user intention set includes all the intentions provided by the user.

[0104] The classification unit 402 can classify the user intentions in the user intention set to obtain the classified user intentions. In some embodiments, the user intentions can be classified from three aspects: business scenario, user type and business attribute.

[0105] The conversion unit 403 can convert the classified user intent into a network policy. In some embodiments, each classification has a corresponding policy library, and the corresponding policy library can be determined according to the classified user intent, and the classified user intent is converted into a network policy based on the determined policy library. In some embodiments, different classifications have corresponding converters, and the corresponding converters can be determined according to the user intent to convert the classified user intent into a network policy, and the policy libraries based on different converters are different.

[0106] The verification unit 404 can verify the network policy by executing the network policy in the network environment and obtain a verification result. In some embodiments, the verification result is for each user policy.

[0107] The decision unit 405 can determine whether the network policy corresponding to the verification result is passed according to the verification result. For the network policy that passes the verification, it is determined to adopt the network policy; for the policy that fails the verification, it needs to be modified. In some embodiments, the conversion unit 403 can be informed to re-convert the corresponding intention into the network policy.

[0108] Such a device classifies the intent before converting it into a network policy, thereby reducing the amount of data processed by retrieval and polling in the policy library, reducing time complexity, and improving processing efficiency; reducing the probability of erroneous conversion and improving accuracy.

[0109] In some embodiments, Figure 4A As shown, the user intention implementation device in the intent-driven network may also include an intent management unit 406, which is capable of verifying the user intention in the user intention set obtained by the intent acquisition unit; feeding back the user intention that fails the verification to the user so that the user can modify it; and sending the user intention that passes the verification to the classification unit.

[0110] Such a device can first verify and organize the user's intention to ensure that the subsequent classification and conversion work can proceed smoothly.

[0111] In some embodiments, the operation process of the user intention realization device in the intention-driven network can be as follows: Figure 4B shown.

[0112] 1. Intent management: After receiving user intents from the business layer, the intent orchestration layer needs to verify and classify these intents. Qualified intents will be further submitted to the classification unit for analysis and processing. Unqualified intents will be fed back to the user by the intent management unit for further modification and adjustment.

[0113] 2. Conversion: According to the analysis results of the classification unit, they are assigned to the corresponding converters in the conversion unit and configured and issued, and the results are output to the verification unit.

[0114] 3. Verification: The verification unit has two verification functions:

[0115] The first is configuration feasibility verification. The configuration obtained in the translation process needs to be verified based on the real-time network status to see whether it can be executed in the actual network environment, and relevant information should be fed back to the decision-making unit.

[0116] The second is validity verification. When the configuration is actually executed, the network status will change. At this time, it is necessary to verify whether the execution of the configuration makes the network status change as expected, and feedback the verification results to the decision module. If the verification fails, it means that the intent layer needs to perform "intent-configuration" orchestration again.

[0117] 4. Decision-making: Responsible for the overall control of network status and configuration, it not only processes the data transmitted by the verification unit, but also assists the conversion unit in converting intent. The decision-making unit supports the experience and knowledge of upper-level administrators to help with decision-making, and can also use artificial intelligence to assist in decision-making. When the user's intention is not realized or the network status is abnormal, the decision-making unit needs to make optimization or remedial measures, and notify the conversion unit to respond, and then translate the intention to configuration again.

[0118] The intent execution layer mainly includes the configuration execution of intent and the perception and feedback of network status. The perception information collection function supports the collection of network topology information, network traffic information, business paths, etc. When the configuration is actually executed, feedback information is output to the verification unit.

[0119] Such a device can finely control the generation of intent strategies. According to the key information extracted from the user's intent, it can manage the corresponding classification rules through intent classification and add identification recognition, so that it can more accurately retrieve the corresponding policy library, generate configuration policies and issue them, and improve system reliability. It can effectively improve system efficiency, overcome the defects of high time complexity and inability to guarantee user experience caused by retrieving and extracting key information of intent through large amounts of data polling or enumeration methods in the policy library, effectively identify intent requests based on business scenarios, user types, and business attributes, and perform precise intent conversion by the corresponding intent converter, reducing the interaction frequency between intent verification and intent conversion, and effectively improving system efficiency.

[0120] Schematic diagrams of some embodiments of the classification unit in the user intent implementation device in the intent-driven network of the present disclosure are as follows Figure 5 shown.

[0121] Function constraint 503 can determine the intended user type and service attributes, including two constraint determination modules: feedback and life cycle.

[0122] The classification rule 501 can identify the corresponding business scenarios, such as operator network, DC network and enterprise network, according to the user-id of the user input intention; and identify the corresponding user type and business attributes according to the judgment of the functional constraint. User types include network administrators, operation operators, APP developers, end users, etc.; business attributes include network optimization, connectivity requests, QoS (Quality of Service), security requests, resource requests, device configuration, etc. Different user types have different feedback requirements for intents, and different business attributes have different requirements for the life cycle of intents. According to whether there is a life cycle, it can be divided into two types of intents: intents with a life cycle and intents without a life cycle.

[0123] The attribute collector 502 can extract key fields of the input intent and input them into the classification rule generator to retrieve corresponding information.

[0124] The classification identifier 504 can authenticate the identifier in the classification rule and output it to the corresponding converter module for configuration and distribution according to the combination of various added identifier headers.

[0125] Such a device effectively identifies intent requests based on business scenarios, user types, and business attributes, and converts them into corresponding intent converters for precise intent execution, reducing the frequency of interaction between intent verification and intent conversion, and effectively improving system efficiency.

[0126] A structural diagram of an embodiment of a device for realizing user intention in an intention-driven network disclosed in the present invention is shown in FIG. Figure 6As shown. The device for realizing user intention in an intent-driven network includes a memory 601 and a processor 602. Wherein: the memory 601 can be a disk, a flash memory or any other non-volatile storage medium. The memory is used to store instructions in the corresponding embodiment of the method for realizing user intention in an intent-driven network described above. The processor 602 is coupled to the memory 601 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 602 is used to execute instructions stored in the memory, which can reduce time complexity, improve processing efficiency, reduce the probability of erroneous conversion, and improve accuracy.

[0127] In one embodiment, it is also possible to Figure 7 As shown, the user intention realization device 700 in the intention-driven network includes a memory 701 and a processor 702. The processor 702 is coupled to the memory 701 via a BUS bus 703. The user intention realization device 700 in the intention-driven network can also be connected to an external storage device 705 through a storage interface 704 to call external data, and can also be connected to a network or another computer system (not shown) through a network interface 706. No detailed introduction is given here.

[0128] In this embodiment, by storing data instructions in a memory and then processing the instructions through a processor, time complexity can be reduced, processing efficiency can be improved, the probability of erroneous conversion can be reduced, and accuracy can be improved.

[0129] In another embodiment, a computer-readable storage medium stores thereon computer program instructions, which, when executed by a processor, implement the steps of the method in the corresponding embodiment of the method for implementing the user intent in the intent-driven network. It should be understood by those skilled in the art that the embodiments of the present disclosure may be provided as methods, devices, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0130] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0131] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0132] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0133] So far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0134] The method and apparatus of the present disclosure may be implemented in many ways. For example, the method and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present disclosure rather than to limit it. Although the present disclosure has been described in detail with reference to the preferred embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present disclosure can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present disclosure, which should be included in the scope of the technical solution for protection of the present disclosure.

Claims

1. A method for realizing user intent in an intent-driven network, comprising: Get the user intent set; Classifying the user intentions in the user intention set and obtaining the classified user intentions, including: determining the business scenario, user type and business attributes of each user intention in the user intention set, wherein the business scenario includes the network environment in which the user is located, the network environment includes an operator network, a data center DC network or an enterprise network, the user type includes managers or operation operators who need to obtain network status feedback, and ordinary users or developers who do not need to obtain network status feedback, the business attributes include intentions without a life cycle in which the intention is completed after the specified operation is successfully executed, and intentions with a life cycle, once the intention with a life cycle is successfully activated and deployed, the system will keep all related intentions active until they are deactivated or deleted, and the user intention message includes a business scenario identifier of the corresponding business scenario, a user type identifier of the corresponding type and a corresponding business attribute identifier; Converting the classified user intention into a network policy includes: determining the intended business scenario, user type and business attribute according to the business scenario identifier, the user type identifier and the business attribute identifier, and determining a corresponding policy library so that the converter module corresponding to the policy library can configure and issue the policy; Verifying the network policy by executing the network policy in a network environment and obtaining a verification result; The network policy is confirmed or modified according to the verification result until the user's intention is achieved.

2. The method according to claim 1, wherein: The business scenarios for determining each user intent in the user intent set include: Obtaining an identifier of a user providing the user intent; Determine the network environment where the user is located based on the user's identification.

3. The method according to claim 1, wherein: Determine the user type for each user intent in the user intent set includes: Extract key fields of user intent through attribute collector; The function determiner determines whether the user intention is feedback-oriented according to the key field, wherein the key field has a corresponding relationship with whether the user intention is feedback-oriented; If the user intention is not feedback-oriented, the user type is determined to be a normal user; If the user's intention is feedback-oriented, the user type is determined to be an administrator.

4. The method according to claim 1, wherein: Determine the business attributes of each user intent in the user intent set including: Extract key fields of user intent through attribute collector; The function determiner determines whether the user intention has a life cycle according to the key field; If the user intent has a life cycle, then the business attribute is determined to be an intent with a life cycle; If the user intent does not have a life cycle, the business attribute is determined to be an intent without a life cycle.

5. The method according to claim 1, wherein: The verifying the network policy by executing the network policy in the network environment and obtaining the verification result comprises: executing the network policy in the network environment; Generate an executability verification result based on whether the network policy is successfully executed; Generate a validity verification result according to whether the network status change meets the expectation of the user's intention; The verification result includes the executability verification result and the validity verification result.

6. The method according to claim 1, wherein: Confirming or modifying the network policy according to the verification result until the user's intention is achieved includes: According to the verification result, based on artificial intelligence or administrator instructions, filter out network policies that fail the verification; Determine the classified user intent corresponding to the network policy that failed the verification, and re-execute the operation of converting the classified user intent into the network policy until the verification passes.

7. The method according to claim 1, further comprising: After obtaining the user intention set, verifying the user intention in the user intention set; Feedback the user's intention that fails the verification to the user so that the user can modify it; An operation of classifying the user intention set is performed on the user intention that passes the verification.

8. A user intention realization device in an intent-driven network, comprising: An intention acquisition unit, configured to acquire a user intention set; A classification unit is configured to classify the user intention set and obtain the classified user intention, including: determining the business scenario, user type and business attribute of each user intention in the user intention set, wherein the business scenario includes the network environment in which the user is located, the network environment includes an operator network, a data center DC network or an enterprise network, the user type includes a manager or an operation operator who needs to obtain network status feedback, and an ordinary user or a developer who does not need to obtain network status feedback, the business attribute includes an intention without a life cycle in which the intention is completed after the specified operation is successfully executed, and an intention with a life cycle, once the intention with a life cycle is successfully activated and deployed, the system will keep all related intentions active until they are deactivated or deleted, and the user intention message includes a business scenario identifier of the corresponding business scenario, a user type identifier of the corresponding type and a corresponding business attribute identifier; A conversion unit is configured to convert the classified user intent into a network policy, including: determining the intended business scenario, user type and business attribute according to the business scenario identifier, the user type identifier and the business attribute identifier, and determining a corresponding policy library so that the converter module corresponding to the policy library can configure and issue it; A verification unit, configured to verify the network policy by executing the network policy in a network environment, and obtain a verification result; The decision-making unit is configured to confirm or modify the network policy according to the verification result until the user intention is achieved.

9. The apparatus according to claim 8, further comprising: an intention management unit, configured to verify the user intention in the user intention set acquired by the intention acquisition unit; The user intention that fails the verification is fed back to the user so that the user can modify it; the user intention that passes the verification is sent to the classification unit.

10. A device for realizing user intention in an intention-driven network, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to execute the method according to any one of claims 1 to 7 based on instructions stored in the memory.

11. A computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images