Method and network node for mapping an intent or service request to one or more resources

Abstract

The disclosure relates to a computer implemented method, and network node, for mapping an intent or service request to one or more resources. The method comprises obtaining a service graph and requirements, corresponding to the intent or service request. The method comprise mapping the requirements to key performance indicators (KPIs) associated with services in the service graph. The method comprises deploying the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

Description

METHOD AND NETWORK NODE FOR MAPPING AN INTENT OR SERVICE REQUEST TO ONE OR MORE RESOURCESTECHNICAL FIELD

[0001] The present disclosure relates to the mapping of intent-aware expectations to services using Artificial Intelligence (AI) / Machine Learning (ML) in the cloud.BACKGROUND

[0002] Intent-based networking is a new paradigm that aims at easing the management of cloud systems and bringing intelligence to traditional networks. The European Telecommunications Standards Institute (ETSI) Experiential Networked Intelligence (ENI) defines intent as a set of operational goals that a network should meet and the outcomes that a network is supposed to deliver, defined in a declarative manner without specifying how to achieve or how to implement them. An example of an Intent could be: “A web service to handle 1000 transactions per minute and provide the highest levels of security and reliability, with a latency no longer than 3 seconds”. In general, intents are expressed as high-level requirements in terms of what is required and not how to achieve it (without low-level deployment and infrastructure details). Therefore, a translation between what is required and how to achieve the intent is needed to determine the appropriate requirements for the services that will fulfill the user expectations.

[0003] Given the high-level description of an intent, there exist solutions to decompose an input intent into a set of specific services (e.g., Virtualized Network Functions (VNFs), microservices, etc.) and chain them according to their dependencies. Those identified services can be deployed without specifying the expectations of each service (e.g., number of requests per node, maximum delay for database, delay jitter, loss rate, etc.). In other words, the expectations of those services are defined on the fly when running them according to the availability of resources, failure rates, etc. Such an approach can’t guarantee that these services achieve their goals and satisfy the expectations defined in the input intent. This is because the capabilities of each service vary over time due to many factors related to infrastructure, services dependencies, etc. which can result in violation of the intent after being deployed.SUMMARY

[0004] Therefore, the expectations of each service in the intent-based service graph should be determined to ensure the successful deployment of the intent and how much each service contributes to the requirements in order to reduce the risks of intent violations on the underlying cloud system. In addition, without knowing the expectations of each service, it is very challenging to identify the required number of resources to be allocated to each service to deliver the intent requirements, which can result in resources over / under-allocation. Consequently, it is important to understand how a declarative intent can dictate the expectations of the required services to be deployed to guarantee a successful deployment of the intent.

[0005] There is provided a computer implemented method for mapping an intent or service request to one or more resources. The method comprises obtaining a service graph and requirements, corresponding to the intent or service request. The method comprises mapping the requirements to key performance indicators (KPIs) associated with services in the service graph. The method comprises deploying the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

[0006] There is provided a network node, operative to map an intent or service request to one or more resources. The network node comprises processing circuits and a memory. The memory contains instructions executable by the processing circuits whereby the network node is operative to obtain a service graph and requirements, corresponding to the intent or service request. The network node is operative to map the requirements to key performance indicators (KPIs) associated with services in the service graph. The network node is operative to deploy the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

[0007] There is provided a non-transitory computer readable media having stored thereon instructions for mapping an intent or service request to one or more resources. The instructions comprise obtaining a service graph and requirements, corresponding to the intent or service request. The instructions comprise mapping the requirements to key performance indicators (KPIs) associated with services in the service graph. The instructions comprise deploying the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

[0008] The method and network node provided herein present improvements to the way how mapping of an intent or service request to one or more resources operate.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure l is a block diagram of an example system for service expectations mapping.

[0010] Figure 2a is a schematic illustration of an example translation of an input intent to a service graph.

[0011] Figures 2b, 2c and 2d are schematic illustration of examples mappings between intent requirements and service expectations (key performance indicators).

[0012] Figure 3 is a schematic illustration of an example 5G core service graph, including services and relationships.

[0013] Figure 4 is a flowchart of an example method for mapping intent-aware expectations to services.

[0014] Figure 5 is a flowchart of a method for mapping an intent or service request to one or more resources.

[0015] Figure 6 is a schematic illustration of a hardware in which steps and / or method described herein can be executed.

[0016] Figure 7 is a schematic illustration of a virtualization environment in which the different steps and hardware components described herein can be deployed.DETAILED DESCRIPTION

[0017] Various features will now be described with reference to the drawings to fully convey the scope of the disclosure to those skilled in the art.

[0018] Sequences of actions or functions may be used within this disclosure. It should be recognized that some functions or actions, in some contexts, could be performed by specialized circuits, by program instructions being executed by one or more processors, or by a combination of both.

[0019] Further, computer readable carrier or carrier wave may contain an appropriate set of computer instructions that would cause a processor to carry out the techniques described herein.

[0020] The functions / actions described herein may occur out of the order noted in the sequence of actions or simultaneously. Furthermore, in some illustrations, some blocks, functions or actions may be optional and may or may not be executed; these are generally illustrated with dashed lines.

[0021] At least some aspects of the techniques described herein may be implemented using artificial intelligence, which comprises a variety of techniques as would beapparent to a person skilled in the art, including machine learning techniques. Machine learning techniques include Neural Network (NN), or Artificial Neural Network (ANN), and both terms may be used interchangeably herein. In some contexts, an Artificial Neural Network could include biological portions.

[0022] Further, looking forward, in a virtual world (e.g., the metaverse, digital twins, etc.), the techniques described herein could be applied in relevant virtual scenarios.

[0023] Herein, a system and a method that can map intent-aware expectations to services using AI / ML technologies is described. The mapped service expectations determine what key performance indicators (KPIs) should be configured for each service such that the intent requirements are satisfied.

[0024] The proposed system takes as input the description of the high-level intent or a service request, along with its translation in terms of service graph and requirements, and a service catalog that describes the services (services can be Network functions (NFs), VNFs, microservices, etc.).

[0025] The system then maps the intent requirements to the services defined in the input service graph by mapping the requirements defined in the intent to a set of expectations represented by KPIs to be configured for each service. This is done by identifying correlation between services in a service graph, correlation between the different intent requirements, and correlation between the list of KPIs identified for each service and the requirements of the intent.

[0026] The output of the proposed system is a list of expectations represented by KPIs mapped to each service to be configured.

[0027] The system provides an automated solution having several advantages, including:- the system provides a repository to save request or intent requirements and the corresponding mapped expectations of each service;- the system considers soft and hard requirements and requirements priorities while mapping expectations for services;- the system finds possible conflicts between hard and soft requirements while mapping expectations for services;- the system reduces human intervention in checking and identifying appropriate expectations in terms of KPIs to be configured in a service graph; and- the system improves resources allocation to the services defined in the service graph to fulfill the intent requirements (avoiding under / over utilization).

[0028] The solution proposes mapping requirements defined in an input intent to: services defined in the service graph required to deploy the input intent in automated manner; and a set of expectations of each service (e.g., NF, microservice, etc.) in a service graph in an automated manner.

[0029] The expectations define a KPI or a list of KPIs to be configured for each service to meet the intent requirements. In currently available solutions, these KPIs are not specified, and they are monitored after the intent-based service graph is deployed. Alternatively, they can be defined through human experts who should have a deep knowledge about the requirements, service graph and the service expectations (KPIs). When using previous solutions to configure the services without having the appropriate mapping between intent requirements and service KPIs, such mapping does not guarantee an optimal solution and oftentimes it is expensive (in terms of time and resources).

[0030] Figure 1 presents an overview of an automated service expectation mapping system 100 that can map an input intent 102 (service graph + requirements) into a set of expectations 104 for each service in the service graph. This allows interested entities to find the appropriate expectations’ values to deploy the intent service graph and fulfill the intent requirements.

[0031] The system inputs include the description of the high-level intent provided by an end-user, the translation of the input intent into a service graph that includes a set of services, a set of requirements, and the service catalog. The service graph represents the list of services to be deployed to get the expected intent output. The intent requirements describe the list of requirements that are expected to be met when deploying the service graph from start to end. The service catalog includes a description of services (e.g., VNFs, microservices, etc.) that could be deployed for the running cloud applications including their main properties (e.g., name, type, priority, dependencies, etc.).

[0032] The system output consists of the services’ expectations represented by KPIs to be configured for each service. These KPIs includes hard KPIs, that are required to be satisfied at all times, soft KPIs, that should be satisfied as much as possible, and conflicting KPIs that need to be assigned priorities according to the requirements defined in the intent request. Herein, the words “KPIs” and “constraints may be used interchangeably.

[0033] The proposed system takes the list of intent requirements and maps them to their corresponding services within the service graph. It identifies the list of expectations (in terms of KPIs) of each service and then finds a correlation between those expectations and the input requirements. It also identifies correlations between services in a service graph and, correlations between requirements defined in the intent request. Next, it analyzes the mapping of these requirements into expectations for each service and finds the hard KPIs, the soft KPIs, and the conflicting ones that can be given as hints when configuring them by corresponding entities.

[0034] Turning to figure 2a, a first example is presented to allow for a better understanding of the mapping of an input intent (in terms of requirements) to a set of expectations for the services to be deployed to fulfill the intent requirements. In the example, the input intent 202 is: “A web service with high levels of security and availability that could process 10000 requests per minutes, request duration = 10s, with latency no longer than 3 seconds”. It is assumed that the input intent is translated to the corresponding service graph shown 204, which shows examples of requirements that are identified. In this example, the requirements could be Latency (Rl), Maximum number of requests (R2), and Request Duration (R3), etc.

[0035] In Figures 2b-2d an example of mapping between the identified requirements and the services in the service graph is presented. Figure 2b shows that the latency requirement (Rl) has a direct impact on SI, S2, S4, and S5 services. Figure 2c shown that maximum number of requests requirement (R2) has an impact on SI, S3 and S5. Figure 2d shows that the request duration requirement (R3) has an impact on SI and S3.

[0036] The following also illustrates an example of mapping between the identified requirements for each service and their corresponding KPIs.1- Latency = requirement, R1KPIs = {S1 [KPI1 : throughput, KPI2: delay], S2[KPI2: throughput], S4[KPI1 : throughput, KPI2: delay], S5[KPI1: throughput, KPI6: request_success_rate, KPI7: frame_rate]}2- maximum number requests = requirement, R2KPIs = {S1 [KPI2: delay, KPI3: packet loss], S3[KPI2: delay, KPI3: packet loss], S5[KPI1: throughput , KPI6: request_success_rate, KPI7: frame_rate]}3- request duration = requirement, R3KPIs = {S1 [KPI2: delay, KPI3: packet loss], S3[KPI3: packet loss, KPI7: frame_rate]}

[0037] Figure 3 illustrates the fifth generation (5G) core and another example is provided in that context.

[0038] In the context of 5G core, an example of intent could be: “A 5G core able to support 5k subscribers with high quality of experience, a latency less than 10ms and a handover success rate greater than 95%”.

[0039] The solution provided herein is able to map the intent requirements defined in this example (number of subscribers, quality of experience, latency, handover success rate). While latency and quality of experience is affecting all service defined the 5G service graph given in Figure 3, KPIs to be configured for each of those service should be identified. Furthermore, the handover success rate is affecting some of the service in the service graph (e.g., Access and Mobility Management Function (AMF), Session Management Function (SMF)). Accordingly, the solution gets the relevant KPIs of those services that have a direct impact on the intent requirements handover success rate.

[0040] Figure 4 is a flowchart illustrating of a method 400 for mapping intent-aware expectations to services.

[0041] STEP 401 : Given a list of services in the input service graph, and requirements, the method starts by identifying group of services contributing to each requirement. These can be given either as input from the system operator, or clustering techniques can be used to identify the related services.

[0042] STEP 402: The method then determines the list of expectations in terms of KPIs that can be mapped for each service using the services catalogues and profiles 420. In one embodiment, the services can be implemented as VNFs. Accordingly, these catalogues and profiles can represent the repository of all on-boarded VNF Packages, supporting the creation and management of the VNF Package (VNF Descriptor (VNFD), software images, manifest files, etc.).

[0043] A VNFD includes the deployment and operational behavior of a VNF. The VNFD defines VNF properties including resources needed (amount and type of virtual compute, storage, networking), software metadata, connectivity, lifecycle management behavior (e.g., scaling, instantiation), supported lifecycle management operations and their configuration, supported VNF specific parameters, affinity / anti- affinity rules, etc.

[0044] VNF profiles present data about the VNF performance with different possible KPIs. These profiles can be either built exhaustively by testing all possible KPIs orcan be automated by prediction methods that can be used to predict the performance metrics of the VNFs from the resource allocation and workload metrics.

[0045] STEP 403: The method then performs a correlation analysis between the requirements defined in the intent. This is done by measuring the strength of the relationship between the intent requirements. The goal of this step is to identify the relations between the requirements, meaning which requirement(s) has / have an impact on a given requirement. One way to measure such correlation could be to use Spearman and / or Pearson correlation metrics, which can be found here: statistics.laerd.com / statistical-guides / spearmans-rank-order-correlation-statistical- guide. php. Spearman correlation and Pearson correlation both are statistical measures that assess the strength and direction of the relationship between two variables. The Pearson correlation coefficient, also known as Pearson's r, measures the linear relationship between two continuous variables. It quantifies the degree to which the variables are linearly related and ranges between -1 and 1. The Spearman correlation coefficient, also known as Spearman's rho (p), assesses the monotonic relationship between two variables. Unlike Pearson correlation, Spearman's rho does not assume a linear relationship between the variables, making it suitable for analyzing relationships that might be nonlinear or non-normally distributed.

[0046] To configure the correlation metrics, thresholds can be used to determine whether the two variables are correlated or not, those thresholds can be an input from an end-user, can be predefined, or could be computed / adjusted dynamically.

[0047] Once the correlation measures are identified, the method ranks the values of correlation measures, selects the correlation with the highest score, and outputs the different lists (one list per requirement).<Req_i, [correlated Reqs]>

[0048] STEP 404: For each requirement, the method gets the group of corresponding services, identified in step 401, that contributes to this requirement. The list of service contributing to each requirement can be given either as input from the system operator, or can be obtained by clustering techniques.

[0049] STEP 405: For each service, given the list of its identified KPIs, the relationships between these KPIs and the intent requirements are identified. This step could be done by using data from the VNF catalogues and profiles to perform a correlation analysis between the KPIs (e.g., response time) and their requirements (e.g., latency). Similar to step 403, this correlation can be measured using Spearmanand / or Pearson correlation metrics. Once the correlation measures are identified, the method ranks the values of correlation measures, selects the correlation with highest score, and outputs the different lists (one list per service). For each identified correlation between KPIs and a given service, it is saved to be used later in next steps.< Service J , { Req i : [KPIs]}>

[0050] STEP 406: In this step, the method selects and maps the generated KPIs, that are correlated to the given requirements (Req i), to their corresponding services / VNFs. These KPIs are mapped as “hard constraints”. To select the correlated KPIs and consider them as hard constraints, the method compares the degree of relationship between the KPIs (using Spearman / Pearson) and compare them to the threshold to consider a KPI as a hard constraint. This threshold is given by the enduser.[ServiceJ, {Req i: hard constraints [selected KPIs]}]

[0051] STEP 407: Next, the method saves the KPIs generated for each requirement to be used in later steps <Req_i, [KPIs]>, and then, goes to the next requirement and repeats steps 404 to 407.

[0052] STEP 408: After identifying the list of hard constraints (KPIs) of each requirement (which was done in step 406) the method gets the list of correlated requirements for each requirement. This list was identified in step 403.

[0053] STEP 409: Once the correlated requirements, for each requirement, are identified, the list of KPIs impacting these correlated requirements are mapped to the given service as soft constraints (KPIs). The list of KPIs impacting the correlated requirements is obtained from the <Req_i, [KPIs]> list saved in step 407.[Service, {Req i: hard constraints [selected KPIs]}, {Req J : soft constraints [selected KPIs]}]

[0054] STEP 410: Finally, after mapping hard and soft constraints (KPIs) for each service, the system identifies potential conflicts between these KPIs. Conflicts between KPI can be identified using different approaches including correlation analysis between the KPIs (similar to the approach to identify the strength between the KPIs), clustering method, regression method, time series analysis etc. These conflicts can be given as hints to interested entities to properly configure the KPIs for each service.[Service, {Req i: hard constraints [selected KPIs]}, {ReqJ : soft constraints [selected KPIs]}, {conflicting KPIs}]

[0055] Turning to figure 5, there is provided a computer implemented method 500 for mapping an intent or service request to one or more resources. The method comprises obtaining, step 502, a service graph and requirements, corresponding to the intent or service request. The method comprises mapping, step 504, the requirements to key performance indicators (KPIs) associated with services in the service graph. The method comprises deploying, step 526, the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

[0056] Mapping the requirements to the KPIs may comprise identifying, step 506, one or more services in the service graph contributing to the requirements and identifying, step 508, a list of KPIs for each service of the one or more services contributing to the requirements.

[0057] Identifying the one or more services in the service graph contributing to the requirements may comprise applying a clustering technique to identify related services.

[0058] Identifying the list of KPIs for each service of the one or more services may comprise using a services catalogues and associated service profiles. The service catalogue may be a repository of on-boarded Virtual Network Functions (VNFs), and the associated service profiles are VNF Packages including VNF Descriptors (VNFDs).

[0059] The method 500 may further comprise computing, step 510, correlations between the requirements and generating, 512, a first list of requirement correlations for each service.

[0060] The method 500 may further comprising, for each requirement, identifying, step 514, a group of services contributing to the requirement.

[0061] The method 500 may further comprise, for each service of the group of services contributing to the requirement, computing, step 516, correlations between the KPIs of the service and the requirement and generating, step 518, a second list of requirement correlations for each service. Computing the correlations may comprise using Spearman or Pearson correlation metrics for measuring a strength of a relationship between correlated elements.

[0062] The method 500 may further comprise ranking, step 520, the requirement correlations of the second list and identifying KPIs impacting top requirement correlations of the second list as hard constraints and ranking, step 522, the requirement correlations of the first list and identifying KPIs impacting toprequirement correlations of the first list as soft constraints. The method may further comprise identifying, step 524, conflicting KPIs for each service.

[0063] It should be noted that methods and steps described herein are, generally, computer implemented methods and steps. The term computer may be interpreted as having different meanings, such as explained next, for example.

[0064] Referring to figure 6, there is provided a network node (HW) 601, in which functions and steps described herein can be implemented.

[0065] The network node 601 (which may go beyond what is illustrated in figure 6), may be a user device, such as a smartphone, tablet, computer, wearable, connected vehicle, including but not limited to bicycle, car, truck, plane, drone, etc.

[0066] The network node 601 may be a server, network node, radio base station, or other computing device which may be part of a cloud computing system, edge computing system, or which may be a standalone device.

[0067] The network node 601 may be an internet of things (loT) device. The loT device may be a sensor, e.g., a thermometer or any detector for detecting environmental conditions, a camera, medical sensor, fitness tracker, etc. The loT device may be a gaming console, appliance e.g., refrigerator, printer, smart thermostat, smart TV, voice control device, lighting appliance, smart doorbell, alarm system, home robot, etc.

[0068] The system proposed herein defines the most appropriate KPIs (hard and soft constraints, and conflicting ones) for each service in a way that the mapped KPIs reflect and fulfill intent requirements. This will help the system operators:- to properly set required resources for each service to fulfill the intent requirements (without over / under utilizing resources);- to properly configure the values of those KPIs using existing solutions (e.g., INSpIRE [E. J. Scheid et al., "INSpIRE: Integrated NFV-based Intent Refinement Environment," 2017 IFIP / IEEE Symposium on Integrated Network and Service Management (IM), 2017, pp. 186-194]).

[0069] The network node 601 comprises processing circuitry 603 and memory 605. The memory 605 can contain instructions executable by the processing circuitry 603 whereby functions and steps described herein may be executed to provide any of the relevant features and benefits disclosed herein.

[0070] The network node 601 may also include non-transitory, persistent, machine- readable storage media 607 having stored therein software and / or instruction 609executable by the processing circuitry 603 to execute functions and steps described herein. The network node may also include network interface(s) and a power source.

[0071] The instructions 609 may include a computer program for configuring the processing circuitry 603. The computer program may be stored in a physical memory local to the device, which can be removable, or it could alternatively, or in part, be stored in the cloud. The computer program may also be embodied in a carrier such as an electronic signal, optical signal, radio signal, or computer readable storage medium.

[0072] Referring to figure 7, there is provided a virtualization environment 700 in which functions and steps described herein can be implemented.

[0073] The virtualization environment 700 (which may go beyond what is illustrated in figure 7), may comprise systems, networks, servers, nodes, devices, etc., that are in communication with each other either through wire or wirelessly, e.g., through a network interface component (NIC) comprising physical network interface(s). Some or all of the functions and steps described herein may be implemented as one or more virtual components (e.g., via one or more applications, components, functions, virtual machines, containers, etc.) executing on one or more physical apparatus in one or more networks, systems, environment, etc.

[0074] A virtualization environment provides hardware 701 comprising processing circuitry 703 and memory 705. The memory 705 can contain instructions executable by the processing circuitry 703 whereby functions and steps described herein may be executed to provide any of the relevant features and benefits disclosed herein.

[0075] The hardware 701 may also include non-transitory, persistent, machine- readable storage media 707 having stored therein software and / or instruction 709 executable by the processing circuitry 703 to execute functions and steps described herein.

[0076] The instructions 709 may include a computer program for configuring the processing circuitry 703. The computer program may be stored in a removable memory, such as a portable compact disc, portable digital video disc, or other removable media. The computer program may be stored in a physical memory local to the hardware 701, which can be removable, or it could alternatively, or in part, be stored in the cloud. The computer program may also be embodied in a carrier such as an electronic signal, optical signal, radio signal, or computer readable storage medium.

[0077] Referring again to figures 6 and 7, there is provided a network node 601, 701, operative to map an intent or service request to one or more resources. The network node comprises processing circuitry 603, 703 and a memory 605, 705. The memory 605, 705 contains instructions executable by the processing circuitry 603, 703 whereby the network node is operative to obtain a service graph and requirements, corresponding to the intent or service request. The network node is operative to map the requirements to key performance indicators (KPIs) associated with services in the service graph. The network node is operative to deploy the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

[0078] The network node 601, 701 is further operative to identify one or more services in the service graph contributing to the requirements and identify a list of KPIs for each service of the one or more services contributing to the requirements.

[0079] The network node 601, 701 is further operative to, when identifying the one or more services in the service graph contributing to the requirements, apply a clustering technique to identify related services.

[0080] The network node 601, 701 is further operative to, when identifying the list of KPIs for each service of the one or more services, use a services catalogues and associated service profiles. The service catalogue may be a repository of on-boarded Virtual Network Functions (VNFs), and the associated service profiles are VNF Packages including VNF Descriptors (VNFDs).

[0081] The network node 601, 701 is further operative to compute correlations between the requirements and generate a first list of requirement correlations for each service.

[0082] The network node 601, 701 is further operative to, for each requirement, identify a group of services contributing to the requirement.

[0083] The network node 601, 701 is further operative to, for each service of the group of services contributing to the requirement, compute correlations between the KPIs of the service and the requirement and generate a second list of requirement correlations for each service. The network node is further operative to, when computing the correlations, use Spearman or Pearson correlation metrics for measuring a strength of a relationship between correlated elements.

[0084] The network node 601, 701 is further operative to rank the requirement correlations of the second list and identifying KPIs impacting top requirement correlations of the second list as hard constraints and rank the requirementcorrelations of the first list and identifying KPIs impacting top requirement correlations of the first list as soft constraints. The network node is further operative to identify conflicting KPIs for each service.

[0085] Still referring to figures 6 and 7, there is provided a non-transitory computer readable media 607, 707 having stored thereon instructions 609, 709 for mapping an intent or service request to one or more resources. The instructions 609, 709 comprise obtaining a service graph and requirements, corresponding to the intent or service request. The instructions 609, 709 comprise mapping the requirements to key performance indicators (KPIs) associated with services in the service graph. The instructions 609, 709 comprise deploying the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

[0086] The instructions 609, 709 may further comprise instructions for executing any of the steps described herein.

[0087] Modifications will come to mind to one skilled in the art having the benefit of the teachings presented in the foregoing description and the associated drawings. Therefore, it is to be understood that modifications, such as specific forms other than those described above, are intended to be included within the scope of this disclosure. The previous description is merely illustrative and should not be considered restrictive in any way. The scope sought is given by the appended claims, rather than the preceding description, and all variations and equivalents that fall within the range of the claims are intended to be embraced therein. Although specific terms may be employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

CLAIMS1. A computer implemented method for mapping an intent or service request to one or more resources, comprising: obtaining a service graph and requirements, corresponding to the intent or service request;- mapping the requirements to key performance indicators (KPIs) associated with services in the service graph; and deploying the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

2. The method of claim 1, wherein mapping the requirements to the KPIs comprises:- identifying one or more services in the service graph contributing to the requirements; and- identifying a list of KPIs for each service of the one or more services contributing to the requirements.

3. The method of claim 2, wherein identifying the one or more services in the service graph contributing to the requirements comprises applying a clustering technique to identify related services.

4. The method of claim 2 or 3, wherein identifying the list of KPIs for each service of the one or more services comprises using a services catalogues and associated service profiles.

5. The method of claim 4, wherein the service catalogue is a repository of onboarded Virtual Network Functions (VNFs), and the associated service profiles are VNF Packages including VNF Descriptors (VNFDs).

6. The method of claim 2, further comprising: computing correlations between the requirements; and- generating a first list of requirement correlations for each service.

7. The method of claim 6, further comprising, for each requirement:- identifying a group of services contributing to the requirement.

8. The method of claim 7, further comprising, for each service of the group of services contributing to the requirement: computing correlations between the KPIs of the service and the requirement; and- generating a second list of requirement correlations for each service.

9. The method of claim 6 or 8, wherein computing the correlations comprises using Spearman or Pearson correlation metrics for measuring a strength of a relationship between correlated elements.

10. The method of claim 9, further comprising:- ranking the requirement correlations of the second list and identifying KPIs impacting top requirement correlations of the second list as hard constraints; and- ranking the requirement correlations of the first list and identifying KPIs impacting top requirement correlations of the first list as soft constraints.

11. The method of claim 10, further comprising:- identifying conflicting KPIs for each service.

12. A network node, operative to map an intent or service request to one or more resources, comprising processing circuits and a memory, the memory containing instructions executable by the processing circuits whereby the network node is operative to: obtain a service graph and requirements, corresponding to the intent or service request;- map the requirements to key performance indicators (KPIs) associated with services in the service graph; and deploy the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

13. The network node of claim 12, further operative to:- identify one or more services in the service graph contributing to the requirements; and- identify a list of KPIs for each service of the one or more services contributing to the requirements.

14. The network node of claim 13, further operative to, when identifying the one or more services in the service graph contributing to the requirements, apply a clustering technique to identify related services.

15. The network node of claim 13 or 14, further operative to, when identifying the list of KPIs for each service of the one or more services, use a services catalogues and associated service profiles.

16. The network node of claim 15, wherein the service catalogue is a repository of onboarded Virtual Network Functions (VNFs), and the associated service profiles are VNF Packages including VNF Descriptors (VNFDs).

17. The network node of claim 13, further operative to: compute correlations between the requirements; and- generate a first list of requirement correlations for each service.

18. The network node of claim 17, further operative to, for each requirement:- identify a group of services contributing to the requirement.

19. The network node of claim 18, further operative to, for each service of the group of services contributing to the requirement: compute correlations between the KPIs of the service and the requirement; and- generate a second list of requirement correlations for each service.

20. The network node of claim 17 or 19, further operative to, when computing the correlations, use Spearman or Pearson correlation metrics for measuring a strength of a relationship between correlated elements.

21. The network node of claim 20, further operative to:- rank the requirement correlations of the second list and identifying KPIs impacting top requirement correlations of the second list as hard constraints; and- rank the requirement correlations of the first list and identifying KPIs impacting top requirement correlations of the first list as soft constraints.

22. The network node of claim 21, further operative to:- identify conflicting KPIs for each service.

23. A non-transitory computer readable media having stored thereon instructions for mapping an intent or service request to one or more resources, the instructions comprising: obtaining a service graph and requirements, corresponding to the intent or service request;- mapping the requirements to key performance indicators (KPIs) associated with services in the service graph; and deploying the service graph on an infrastructure by configuring the one or more resources according to the KPIs.

Images