Systems and methods for automated assurance based on AI policies

Abstract

The system includes a processing circuit; and a memory coupled to the processing circuit, the memory configured to store executable instructions that, in response to being executed by the processing circuit, facilitate performance of operations including receiving, by a policy manager, one or more events from a data bus; triggering an existing assurance policy based on the one or more events; triggering a prediction request to an artificial intelligence (AI) / machine learning (ML) application based on the existing assurance policy; receiving, by the policy manager, change recommendations from the AI / ML application; generating, by the policy manager, a new assurance policy based on the change recommendations from the AI / ML application; and triggering, by the policy manager, a corrective action to be taken based on triggering the new assurance policy that implements the change recommendation.

Description

Technical Field

[0001] This description relates to a system for automatic guarantee based on an AI recommendation policy and a method of using the same.

Background Art

[0002] Event-driven architecture (EDA) is a software architecture that facilitates the production, detection, consumption, and reaction to events. An event is a change in state or a labeled annotation based on the log output of an entity in a system. For example, when a consumer purchases an online product, the state of the product changes from "for sale" to "sold". The seller's system architecture treats this state change as an event, and its occurrence is made known to other applications within the architecture.

Summary of the Invention

Problems to be Solved by the Invention

[0003] What is generated, issued, propagated, detected, or consumed are messages called event notifications and non-events, which are state changes that trigger message emission. To report an event that has occurred, an event message is generated and propagated. Nevertheless, the term "event" is often used metaphorically to represent a notification event message. EDA is often designed on a message-driven architecture that includes one (e.g., a message) of the text-based inputs to differentiate how each communication is handled.

Means for Solving the Problems

[0004] In some embodiments, the system includes a processing circuit and a memory connected to the processing circuit, configured to store executable instructions that facilitate the performance of an operation which, depending on what is performed by the processing circuit, involves a policy manager receiving one or more events from a data bus, triggering an existing assurance policy based on one or more events, triggering a prediction request to an artificial intelligence (AI) / (ML) machine learning application based on the existing assurance policy, receiving change recommendations from the AI / ML application, generating a new assurance policy based on the change recommendations from the AI / ML application, and triggering corrective actions to be taken based on the policy manager triggering the new assurance policy that implements the change recommendations.

[0005] In some embodiments, the method performed by the processing circuit includes: the policy manager receiving one or more events from a data bus; triggering an existing assurance policy based on one or more events; triggering a prediction request to an artificial intelligence (AI) / (ML) machine learning application based on the existing assurance policy; the policy manager receiving change recommendations from the AI / ML application; the policy manager generating a new assurance policy based on the change recommendations from the AI / ML application; and the policy manager triggering corrective actions to be taken based on the triggering of the new assurance policy that implements the change recommendations.

[0006] In some embodiments, in a non-temporary tangible computer-readable storage medium storing a computer program, the computer program includes instructions that cause a processing circuit to perform operations including, when executed, receiving one or more events from a data bus, triggering an existing assurance policy based on one or more events, triggering a prediction request to an artificial intelligence (AI) / (ML) machine learning application based on the existing assurance policy, receiving change recommendations from the AI / ML application, generating a new assurance policy based on the change recommendations from the AI / ML application, and triggering corrective actions to be taken based on the new assurance policy implementing the change recommendations. [Brief explanation of the drawing]

[0007] Aspects of this embodiment are best understood from the following detailed description, which is read in conjunction with the accompanying drawings. Following standard industry practice, various features are not depicted to actual size. Dimensions of various features are arbitrarily increased or decreased for clarity of discussion.

[0008] Figure 1 is a block diagram of a correlation and policy engine (CPE) according to several embodiments.

[0009] Figure 2 shows graphical representations of CPE in several embodiments.

[0010] Figure 3 is a data flow diagram of a method for managing physical resource blocks (PRBs) according to several embodiments.

[0011] Figure 4 is a data flow diagram of a method for automatic level scaling according to several embodiments.

[0012] Figure 5 is a data flow diagram of a method for optimizing data center resources according to several embodiments.

[0013] Figure 6 is a high-level functional block diagram of a processor-based system according to several embodiments. [Modes for carrying out the invention]

[0014] The following embodiments include many different examples for implementing different features of the subject. Examples of components, values, operations, materials, arrangements, etc., are described below for the sake of simplicity in these embodiments. These are, needless to say, examples and are not intended to be limiting. Other components, values, operations, materials, arrangements, etc., are also conceivable. For example, the formation of a first feature above or on a second feature in the following description includes embodiments in which the first and second features are formed in direct contact, and further includes embodiments in which an additional feature is formed between the first and second features so that the first and second features are not in direct contact. In addition, these embodiments repeat reference numbers and / or letters in various examples. This repetition is for simplicity and clarity and is not intended to suggest any relationship between the various embodiments and / or configurations discussed.

[0015] Furthermore, spatially relative terms such as "beneath," "below," "lower," "above," and "upper" are available here for ease of description in describing the relationship between one element or feature and other elements or features, as illustrated in the diagram. The spatially relative terms are intended to encompass different orientations of the device during use or operation, in addition to the orientation shown in the diagram. Embodiments of one or more devices may face other directions (rotated 90 degrees or in other directions), and the spatially relative descriptors available here shall be interpreted accordingly.

[0016] The EDA architectural pattern is applied to the design and implementation of applications and systems that transmit event messages between loosely coupled software components and services. An event-driven system typically consists of an event emitter (agent, data source), an event consumer (sink), and an event channel (the medium through which event messages are transmitted from the emitter to the consumer). The event emitter detects, collects, and forwards event messages. The event emitter does not know the consumer of the event message, nor does it know if an event consumer exists, and if a consumer exists, it does not know how the event message will be used or further processed. The event consumer applies a reaction as soon as the event message is presented. The reaction is provided, or not provided, by the event consumer. For example, the event consumer filters event message frames while an event policy performs transformations and generates them to forward the event message frames to other components, or the event consumer provides self-sufficient reactions to such event message frames. An event channel is a conduit through which event message frames are transmitted from an event emitter to an event consumer. In some embodiments, an event consumer becomes an event emitter after receiving an event message frame and forwards the event message frame to other event consumers. The configuration for the correct delivery of event message frames resides within the event channel. The physical implementation of an event channel may be based on components such as message-oriented middleware or point-to-point communication, depending on a more appropriate transaction execution framework (such as a configuration file for establishing the event channel).

[0017] A correlation and policy engine (CPE) is a software application that programmatically understands relationships. CPEs are configured for use in system management tools to collect, normalize, and analyze event data. Event correlation is a technique for understanding the meaning of a large number of events and identifying a small number of significant events within a mass of information. This is achieved by exploring and analyzing the relationships between events. Furthermore, a CPE is a program or process that receives machine-readable policies and applies them to specific problem domains to control the behavior of network resources.

[0018] In programming and software design, an event is a software-recognized change in state (e.g., an action or occurrence) that often arises asynchronously from the external environment handled by the software. Computer event messages are generated or triggered by the system, by the user, or by other event-based methods. Event messages are handled synchronously with the program flow; that is, the software is configured to have one or more dedicated locations (e.g., data sinks) where event messages are handled. Sources of event messages include users interacting with the software through computer peripherals (e.g., by typing on a keyboard). Other sources are hardware devices such as timers. The software is configured to further trigger its own set of event messages within the event channel (e.g., to communicate task completion). Software that changes its behavior in response to event messages is often said to be event-driven, with the goal of being interactive.

[0019] The policy manager is a network assurance policy engine that triggers actions on the northbound system based on whether events received from the southbound system meet defined conditions.

[0020] The policy manager determines (decides) to what extent a service / device is allowed to perform what it is attempting / requesting, and can enforce that decision (enforces it). Some examples of policies include (1) whether a customer can use this service, (2) whether there is sufficient capacity to support this new service, (3) what happens to non-SLA (Service Level Agreement) customers when nodes approach congestion, and (4) whether a service request / activity is a security threat.

[0021] A rules-based system is configured to be used to store and manipulate knowledge in order to interpret information in a useful way. Typically, the term “rules-based system” applies to systems that involve a set of rules created or curated by humans.

[0022] In Software-Defined Networking (SDN), the southbound interface is the OpenFlow protocol specification that enables communication between controllers, switches, and other network nodes with low-level components. It further allows routers to determine the network topology, decide on network flows, and implement requests sent to the router via the northbound interface. The southbound Application Programming Interface (API) enables end users to gain better control over the network and increases the efficiency level of the SDN controller to progress based on real-time demand. In addition, the SDN controller communicates with the forwarding plane to modify the network, causing the SDN controller to move progressively as enterprise calls progress. To configure a network layer that is more responsive to real-time traffic demand, administrators add or remove entries to the internal flow tables of network switches and routers.

[0023] Unlike the southbound API, the northbound interface enables communication between high-level components. Traditional networks use firewalls or load balancers to control data plane behavior, while SDN installs applications that use a controller, and these applications communicate with the controller through the northbound interface. APIs are cleansed by programmers proficient in programming languages, and since this task does not require specialized knowledge, the northbound API facilitates network control and processing innovation or customization for network operators. The northbound interface is an API or protocol that enables low-level network components to communicate with high-level or more central components. On the other hand, the southbound interface enables high-level components to send commands to low-level network components.

[0024] In some embodiments, automated assurance based on artificial intelligence (AI) recommendation policies is discussed. AI is the intelligence realized by machines, as opposed to the natural intelligence shown by animals and humans. AI represents a system that perceives the environment and takes actions to maximize the possibility of achieving goals.

[0025] In other approaches, implementing network assurance is based on manual procedures. Network assurance is based on network key performance indicators (KPIs), and manual tuning has been used to optimize network operations. However, the turnaround time for optimizing network assurance operations remains high (e.g., from several minutes to several days).

[0026] A performance indicator or KPI is a type of performance measurement. KPIs evaluate the success of activities (projects, programs, products, other initiatives, etc.). KPIs provide focus for strategic and operational improvements, generate an analytical basis for decision-making, and help focus attention on the most important things.

[0027] In some embodiments, the policy manager consumes recommendations from AI / ML applications on an on-demand basis according to the situation (e.g., predicted) in the network through an integrated, high-throughput, low-latency platform for handling real-time data feeds. In some embodiments, the policy manager consumes and triggers actions for northbound systems through data related to json (an open standard file format and data exchange format that uses human-readable text to store and transmit data objects composed of attribute-value pairs and arrays (or other serializable values)) events, and the AI / ML application suggests actions based on the learning model. In some embodiments, the overall network assurance (day 2) operations are optimized with the help of AI / ML recommendations streamed on a streaming processing platform (e.g., an event channel) and consumed by a rule engine such as the policy manager. "Day 0" operations are design and procurement, "day 1" operations are installation, provisioning, segmentation, network startup, and "day 2" operations are running the network.

[0028] In some embodiments, eMBB (enhanced mobile broadband) KPI / SLA real-time prediction and physical resource block (PRB) management for AI-driven network assurance are discussed.

[0029] An SLA is a commitment between a service provider and a client. The service aspects include quality, availability, and responsibilities agreed upon between the service provider and the service user. The most common component of an SLA is that the service will be delivered to the customer as agreed in the contract. For example, Internet service providers (ISPs) and telecommunications companies typically include SLAs within the terms of their contracts with customers to define, in plain language, the level of service sold. In this case, an SLA typically includes technical definitions of measurable details such as: Mean Time Between Failures (MTBF), Mean Time to Repair or Mean Time to Recovery (MTTR); identification of the party responsible for reporting failures or paying costs; responsibilities for various data rates; throughput; jitter; or similar measurable details.

[0030] In long-term evolution (LTE is a standard for wireless broadband communication for mobile devices and data terminals based on GSM / EDGE and UMTS / HSPA standards), PRB is described as consisting of 12 consecutive subcarriers per slot (0.5ms) and is the smallest element of resource allocation assigned by the eNodeB (eNB is hardware connected to the cellular network that communicates directly wirelessly with the mobile handset (UE), like the base transceiver station (BTS) in a GSM network) scheduler. EMBB uses 5G as a successor to 4G LTE mobile broadband services, bringing faster connectivity, higher throughput, and more capacity. This is useful for areas with higher traffic, such as stadiums, cities, and concert venues.

[0031] In some embodiments, automatic scaling at the cloud-native network function (CNF), network function (NF), and / or network service (NS) levels using AI predictions is discussed.

[0032] CNFs are software implementations or applications of functions that traditionally run on physical devices but run within containers. Because CNFs operate as containers and use container image registries such as OCI Docker or OS container runtimes, and their lifecycle is managed by Kubernetes, the characteristics of CNFs are containerized microservices that communicate with each other via standardized RESTful APIs, a small performance footprint, the ability to scale laterally, and independence from the guest operating system.

[0033] NF supports (1) Access and Mobility Management Functions (AMF) that support Non-Access Layer (NAS) signaling termination, NAS encryption and integrity protection, registration management, connectivity management, mobility management, access authentication and authorization, and security context management; (2) Session Management Functions (SMF) that support session management (session establishment, modification, release), user equipment (UE) IP address assignment and management, Dynamic Host Configuration Protocol (DHCP) functionality, NAS signaling termination related to session management, downlink (DL) data notification, and traffic steering configuration for user plane functions (UPF) for proper traffic routing; and (3) Packet routing and forwarding, packet inspection, and Quality of Service (QoS) handling, acting as an External Protocol Data Unit (PDU) session point for interconnection to the Data Network (DN) within the RAT (Radio Access Technology). (4) UPF, an anchor point for inter-mobility; (5) Policy control function (PCF) that provides policy rules to control plane (CP) functions, supports access to subscription information for policy decisions in the integrated data repository (UDR); (6) Authentication server function (AUSF) that functions as an authentication server; (7) Integrated data management (UDM) that supports authentication and key agreement (AKA) credential generation, user identification handling, access authorization, and subscription management; (8) Application function (AF) that supports application impact on traffic routing, access to network exposure function (NEF), and interaction with the policy framework for policy control; (9) NEF that supports capability and event exposure, secure provisioning of information from external applications to the 3GPP network, and translation of internal / external information; and (10) Service discovery function.(10) Includes an NF repository function (NRF) that maintains NF profiles and available NF instances, and a network slice selection function (NSSF) that supports the selection of network slice instances to serve to the UE, the determination of acceptable NSSAIs, and the determination of the AMF set that should be used to serve the UE.

[0034] NS is an application that runs above the network application layer, providing data storage, manipulation, presentation, communication, or other capabilities based on application layer network protocols, often implemented using client-server or peer-to-peer architectures. Each service is typically provided by a server component running on one or more computers (often a dedicated server computer providing multiple services) and accessed over the network by client components running on other devices.

[0035] In some embodiments, automated resource optimization across multiple data centers (DCs) using AI predictions is discussed.

[0036] A data center is a building, a dedicated space within a building, or a group of buildings used to house computer systems and related components such as communication and storage systems. Because information technology (IT) operations are critical to business continuity, data centers include redundant or backup components and infrastructure for power supplies, data communication connectivity, environmental control (e.g., air conditioning, fire protection), and various security devices.

[0037] In some embodiments where eMBB slice KPI / SLA real-time forecasting and PRB management are implemented, a virtualized distributed unit (vDU) pushes Performance Monitor (PM) metrics to an Observability Framework (OBF) application once per minute. The VDU runs as a virtual network function (VNF) within a Network Function Virtualization (NFV) infrastructure. The OBF application collects telemetry data from the NF, enabling the use of AI / ML to optimize and operate the 5G network. The OBF application provides near real-time visibility into the performance and operation of the CNF. This collected data is used to optimize the network through a closed-loop automation module that performs procedures to provide automatic scaling and healing while minimizing manual work and reducing errors.

[0038] In some embodiments, an OBF application performs decoding of the received PM data and publishes the PM data to the data bus. The data bus is a communication system that transfers data between components within a computer or between computers. This expression covers the associated hardware components (wires, optical fibers, etc.) and the software, including the communication protocol.

[0039] In some embodiments, the policy manager subscribes to any incoming events on the data bus and data bus sources (e.g., event sources). In some embodiments, an active policy for threshold monitoring in the policy manager is triggered based on one or more incoming events. In some embodiments, the policy manager calculates the UE downlink throughput, expressed as equation (1), based on the triggered policy.

number

[0040] Here, dl is downlink, thpt is throughput, GDRB is guaranteed, thp is throughput, Vol is volume, and D1 is the first data center and packet core gateway (PCB). Thus, the UE throughput volume of the packet core gateway (PCB) is guaranteed. The PCB is the cloud-native user plane traffic processing and gateway function. Throughput represents the success rate of message delivery on a communication channel such as Ethernet or packet radio in a communication network. The data contained in these messages is delivered over physical or logical links or through network nodes. Throughput is typically measured in bits per second (bit / s or bps), sometimes in data packets per second (p / s or pps), or data packets per time slot. System throughput or total throughput is the sum of the data rates delivered to all terminals in the network. Throughput is essentially synonymous with digital bandwidth consumption and is numerically determined by applying queuing theory, where the load on packets per unit of time is expressed as the arrival rate (λ) and the number of packets dropped per unit of time is expressed as the departure rate (μ). The throughput of a communication system is influenced by various factors, including constraints such as the underlying analog physical medium, the available processing power of system components, and end-user behavior.

[0041] In cellular networks, the wireless downlink is the transmission path from the cell site to the mobile phone. Traffic and signaling flowing within the base station subsystem (BSS) and network switching subsystem (NSS) are further identified as uplink and downlink. In computer networks, the downlink is the connection from data communication equipment to data terminal equipment. This is also known as a downstream connection.

[0042] Equation 1 describes how the download throughput of the UE equals the UE throughput volume over time, measured in bits per second. In some embodiments, the UE is a computer or computing system. In addition or alternative, the UE has a liquid crystal display (LCD), light-emitting diode (LED), or organic light-emitting diode (OLED) screen interface that provides a touchscreen interface with digital buttons and a keyboard, or physical buttons along with a physical keyboard. In some embodiments, the UE is connected to the internet and interconnects with other devices. In addition or alternative, the UE integrates an integrated camera, the ability to make and receive voice and video calls, video games, and a global positioning system (GPS) capability. In addition or alternative, the UE runs an operating system (OS) that allows the installation and execution of third-party applications specialized for those capabilities. In some embodiments, the UE is a computer (such as a tablet computer, netbook, digital media player, digital assistant, graphing calculator, handheld game console, handheld personal computer (PC), laptop, mobile internet device (MID), personal digital assistant (PDA), pocket calculator, portable media player, or ultra-mobile PC), a mobile phone (such as a camera phone, feature phone, smartphone, or phablet), a digital camera (such as a digital camcorder or digital still camera (DSC), digital video camera (DVC), or front camera), a pager, a personal navigation device (PND), a wearable computer (such as a calculator watch, smartwatch, head-mounted display, earphone, biometric device), or a smart card.

[0043] In some embodiments, the result of equation (1) is evaluated against a threshold obtained from enriching the slice manager and Cloud Management as a Service (CMaaS) radio resource management (rRm) policy data.

[0044] In some embodiments, the slice manager is responsible for generating network slices and NS subnets. The RRm ensures the efficient use of available network resources. The slice manager is described as a network orchestrator that interfaces with various functions performed by each layer to coherently manage each slice request. The slice manager enables efficient and flexible slice generation that is reconfigured during the slice lifecycle. Operationally, the slice manager is responsible for several tasks that provide more effective coordination between layers, such as end-to-end (E2E) service management, which is the mapping of various service instances (e.g., NSs) expressed in relation to SLA requirements to appropriate network functions that can satisfy the service constraints; virtualization of physical network resources to simplify resource management operations performed to allocate network functions; and slice lifecycle management to dynamically reconfigure each slice to accommodate possible changes in SLA requirements.

[0045] Thresholds include guaranteed throughput (guaThpt), maximum throughput (maxThpt), rRMPolicyMaxRatio, rRMPolicyMinRatio, rRMPolicydedicatedRatio, and download throughput per UE (dLThptPerUE low or dLThptPerUE high).

[0046] The rRM policy includes guidance on splitting wireless resources between multiple slices supported by a cell. The rRM policy ratio is used to specify the percentage of PRB for the corresponding slice, averaged over time. In some embodiments, the sum of the values ​​is less than or equal to 100. The average time is implementation-dependent.

[0047] In some embodiments, the policy manager triggers an AI / ML recommendation request. In some embodiments, the AI / ML application predicts a new threshold and streams the recommendation along with the new PRB count onto the data bus. In some embodiments, the AI / ML recommendation is consumed from the data bus. In some embodiments, a new active policy is based on the AI / ML recommendation. In some embodiments, in response to the triggering of a new active policy, the policy manager triggers the slice manager for a PRB configuration change.

[0048] In some embodiments, CNF / NF / NS level autoscaling using AI predictions is discussed. In some embodiments, UPF N3 DL PM data is sent to OBF applications every minute. The 5G N3 interface plays the role of transmitting user data from the Radio Access Network (RAN) to the User Plane Function (UPF), enabling the generation of both low and high latency services.

[0049] In some embodiments, the OBF application performs enrichment of UPF N3 DL throughput counters in a central inventory. The policy manager includes an enrichment unit that enriches messages arriving with inventory information in order to ensure successful policy execution. In some embodiments, the enrichment unit is configured with a message enrichment cache built by enrichment unit synchronization processing. In a non-limiting example, an incoming event data is missing a field or parameter. The event is then enriched with the help of the inventory to fill in the missing field and parameter so that a decision can be made and a predetermined action can be taken.

[0050] In some embodiments, the policy manager subscribes to the data bus and incoming events. In some embodiments, an active policy for scaling in the policy manager is triggered based on the incoming events. In some embodiments, the policy manager calculates the slice utilization percentage and evaluates whether a threshold has been broken. In a non-limiting example, the scale-in threshold is 40% or less, or the scale-out threshold is 60% or more. Depending on whether a threshold has been broken, the policy manager triggers an AI / ML recommendation request. In some embodiments, the AI / ML application predicts a new threshold and streams the recommendation onto the data bus along with the new scale-in or scale-out threshold. In some embodiments, the AI / ML recommendation is streamed from the data bus and consumed by the policy manager. In some embodiments, a new active policy based on the AI / ML recommendation is generated by the policy manager. In some embodiments, depending on whether a new active policy has been triggered, a slice manager action for a UPF scale-in or scale-out operation is triggered.

[0051] In some embodiments, automated resource optimization with AI predictions across multiple data centers (DCs) is discussed. In some embodiments, user requests for slice optimization are sent to a slice manager via a user interface (UI). In some embodiments, the slice manager streams the requests over a data bus. In some embodiments, a policy manager consumes the requests from the slice manager over the data bus to optimize the slices. In some embodiments, an active policy for optimizing the slices is triggered in the policy manager based on incoming events. In some embodiments, based on the triggered policy, the policy manager triggers an AI / ML recommendation request. In some embodiments, an AI / ML application predicts thresholds and streams the recommendation over the data bus along with "MIGRATE UPF". In some embodiments, the AI / ML recommendation is streamed over the data bus and consumed by the policy manager. In some embodiments, a new active policy is generated in the policy manager based on the AI / ML recommendation. In some embodiments, in response to an incoming event that triggers a new active policy, the policy manager triggers the slice manager to perform a "UPF MIGRATE" operation to a second DC.

[0052] Figure 1 is a block diagram of a correlation and policy engine (CPE) 100 according to several embodiments.

[0053] CPE100 generally includes an event source input block 102, a policy manager block 104, and an action consumer block 106.

[0054] The event source input block 102 includes an event emitter (an agent, a data source, or an event emitter in other suitable embodiments). The event emitter detects, collects, and forwards event messages. The event emitter cannot know the consumers of the event messages, nor can it even know if an event consumer exists, and if a consumer exists, the event emitter cannot know how the event message will be used or further processed.

[0055] Event source 102 includes events from cloud network 108. Cloud network computing is the on-demand availability of computer system resources (particularly data storage (e.g., cloud storage) and computing power) without direct, active management by the user. Large clouds often have distributed functionality across multiple locations, each of which is a data center. Event sources from cloud network 108 are events occurring within the cloud network. In a non-limiting example, one or more incidents may occur within the DC (a building, a dedicated space within a building, or a group of buildings used to house computer systems and related components such as communications and storage systems) of cloud network 108.

[0056] Event source 102 contains events from the 5G core network (CN) 110. The backbone, or CN 110, is part of the computer network that interconnects networks, providing a path for the exchange of information between different local area networks (LANs) or subnetworks. A CN brings together various networks within the same building, within different buildings in a campus environment, or across a wide area. Large enterprises with many locations have a CN that brings together locations, for example, depending on whether server clusters need to be accessed by different departments of the company located in different geographical locations. The portion of the network connectivity (e.g., Ethernet, wireless) that brings these departments together is often referred to as a CN. One example of a CN is the internet backbone. An event source from 5G CN 110 is an event occurring in the 5G CN. In a non-limiting example, one or more incidents occur within a server cluster (a set of servers that work together to be considered a single system (each node is configured, controlled, and scheduled by software to perform the same tasks)) of the 5G CN 110.

[0057] Event source 102 includes events from the 5G Radio Access Network (RAN) 112. The RAN is a part of the mobile communications system. The RAN implements radio access technology. The RAN exists between devices such as mobile phones, computers, or remotely controlled machines and provides connectivity to CNs such as CN 110. Depending on the standard, mobile phones and other radio-connected devices are known by different names such as UE, Terminal Equipment (TE), and Mobile Station (MS). Examples of RAN types include GSM (Global System for Mobile Communications) Radio Access Network, GSM RAN (GRAN), GERAN (essentially the same as GRAN but specifying the inclusion of EDGE packet radio services), Universal Mobile Communications System (UMTS) RAN, UTRAN (UMTS terrestrial RAN), and E-UTRAN (e.g., Long Term Evolution (LTE) High-Speed ​​and Low-Latency Radio Access Network). The event source from 5G RAN 112 is an event occurring in the 5G RAN. In a non-specific example, one or more incidents occur within the TE and / or MS of 5G RAN112.

[0058] Event source 102 includes events from 5G transport network 114. 5G transport network 114 includes fronthaul and backhaul portions.

[0059] The backhaul portion of a network includes intermediate links between the CN (such as CN110) and smaller subnetworks at the network edge. The most common network type in which backhaul is implemented is mobile networks. Mobile network backhaul is also referred to as mobile backhaul, which connects cell sites to the CN. Two methods of mobile backhaul implementation are fiber-based backhaul and wireless point-to-point backhaul. In technical and commercial definitions, backhaul generally represents the side of the network that communicates with the global internet. A middle-mile network may also exist between a customer's own LAN and their exchanges. In some embodiments, this is a local wide area network (WAN) connection.

[0060] The fronthaul network is similar to, but subtly different from, the backhaul network. In Cloud RAN (C-RAN), backhaul data is decoded from the fronthaul network at a centralized controller, from where it is forwarded to the CN. The fronthaul portion of C-RAN includes intermediate links between the centralized radio controller and the radio heads (or masts) at the edge of the cellular network. The event source from the 5G transport network 114 is the events that occur in the 5G transport network 114. In a non-limiting example, one or more incidents may occur within the radio controller or network switch of the 5G transport network 114.

[0061] Policy Manager 104 is a large-scale, real-time, complex event processing (CEP) engine that automates various workflows and network healing operations. CEP consists of a set of concepts and technologies for processing real-time events and extracting information from incoming event streams. CPE 100 processes events based on network assurance policies. Based on given policies and rules, Policy Manager 104 filters, enriches, correlates, and processes events for action.

[0062] The policy manager 104 includes a clean unit 116 that accepts events from the event source block 102, removes undesirable events, and passes the filtered events to the enrich unit 118 for further processing. In some embodiments, these filtered events are forwarded by using a message policy cache built by message policy synchronization processing. During computation, messages are passed between programs or between components of a single program. Message forwarding is a form of communication used in concurrent and parallel computing, object-oriented programming, and channel communication, where communication takes place by sending messages to a recipient. Messages are sent to objects that specify a request for an action.

[0063] The policy manager 104 includes an enrichment unit 118 that enriches messages arriving from the clean unit 116 with inventory information in order to ensure the successful execution of policies. In some embodiments, the enrichment unit 118 consists of a message enrichment cache built by enrichment unit synchronization processing. In a non-limiting example, an incoming event data is missing a field or parameter. The event is then enriched with the help of the inventory to fill in the missing field and parameter so that a decision can be made and a predetermined action can be taken.

[0064] The policy manager 104 includes an evaluation unit 120 that evaluates and processes enriched events arriving from the enrichment unit 118. The evaluation unit 120 is configured to identify the root cause (e.g., what caused or initiated the received event), determine the relevant actions according to a predetermined network assurance policy, and notify the action manager 120 accordingly.

[0065] The policy manager 104 includes a trigger 122 that matches a network assurance policy with an event based on the output of the evaluation unit 120, which identifies the root cause of the received event. The trigger 122 then forwards the matched policy / event to the action consumer 106 to initiate the action workflow.

[0066] Action Consumer 106 includes Ticket Alert 124. Ticket Alert 124 generates an incident or trigger to initiate a workflow action.

[0067] The action consumer 106 includes a trigger workflow 126. In some embodiments, the trigger workflow 126 performs an action based on a user-generated network assurance policy. In some embodiments, the trigger workflow 126 initiates sending notifications. In some embodiments, the trigger workflow 126 initiates a reboot, restart, scale in, scale out, or other appropriate action within the embodiment.

[0068] The action consumer 106 includes a notification action 128. In some embodiments, the notification action 128 is an email, text message, or graphical user interface (GUI) display on a user interface such as user interface 618 (Figure 6) that notifies the policy generation unit and / or network operator that an event has been received, diagnosed, taken, and the result of the action has been taken (e.g., the taken action was successful or failed).

[0069] Figure 2 shows graphic representations of CPE200 in several embodiments.

[0070] In some embodiments, CPE200 is the same as CPE100. In some embodiments, event source 202 is the same as data capture block 102, policy manager 204 is the same as policy manager 104, and action consumer 206 is the same as action manager 106.

[0071] Policy Manager 204 is a large-scale real-time CEP engine that automates various workflows and network healing operations (e.g., repair and / or remediation). Policy Manager 204 processes events based on given network assurance policies and / or rules. Policy Manager 204 filters, enriches, correlates, and processes events for action. Policy Manager 204 provides a framework for supporting CEP capabilities. In some embodiments, computational logic in memory mitigates latency issues. In some embodiments, multi-source event capture covers a wider range of use cases in complex networks and infrastructures. In some embodiments, Policy Manager 204 is configured with a scalable architecture based on business requirements (e.g., new business policies to be implemented). In some embodiments, Policy Manager 204 supports multiple computational logics in near real-time processing such as event "AND", event "OR", event occurrence counting, and mathematical operations on event counters for subsequent events. In a non-limiting example, the calculation logic supports the execution of actions managed by the action manager 230 in response to the "XYZ" event, which is followed by the "ABC" event "AND" ("UVW" event "OR" and "DEF" event) along with the occurrence of 10 "GHI" events. In some embodiments, policy queries are applied to a potentially infinite stream of data. In some embodiments, events are processed immediately. In some embodiments, once the policy manager 204 processes the events to match the sequence, the results are driven directly. In some embodiments, this aspect effectively leads to the policy manager 204 having near real-time capabilities.

[0072] The user and / or network operator generates a policy template using UI208. In some embodiments, UI208 is configured with a GUI that allows the user to view a policy generation template in which they input information for generating a policy. In some embodiments, UI208 is similar to UI618 (Figure 6). In some embodiments, an orchestrator 234 (orchestration is automated configuration, coordination, and management of computer systems and software) provides general policies, artificial intelligence (AI) generated policies, or policies from any external service. The generated policies are sent to a policy manager 210, which relays the generated policies to a database 212.

[0073] The generated policy templates are stored as drafts in database 212. The policy templates are configured to be validated, activated, deactivated, edited, and deleted. Thus, the templates remain stored in database 212 until they are needed and activated in response to user commands.

[0074] The data bus 214 receives data from various sources from the data ingestion block 202, including the cloud platform 216, network application counters 218, container application counters 220, other events via the internet, events via the public cloud 222, and events via the failure and performance system 224. Between the data bus 214 and the cloud platform 216, network application counters 218, and container application counters 220 are the OBF application 234 and the AI / ML application 236.

[0075] The terms “observability” and “monitoring” are often used interchangeably but have different meanings. Observability refers to the ability to obtain the (internal) state of a system based on external output. The goal is to externalize the system state based on sensor data output. Monitoring describes the discipline of collecting metrics and alerts to monitor the health and performance of discrete IT infrastructure components (e.g., servers, storage, network devices). Observability extends the scope of traditional monitoring by collecting different types of data and generating actionable feedback loops using telemetry data. By collecting and correlating data, observability focuses on gaining actionable system-level insights.

[0076] In some embodiments where eMBB slice KPI / SLA real-time forecasting and PRB management are implemented, a virtualized distributed unit (vDU) pushes performance monitor (PM) metrics to OBF application 234 once per minute. The VDU runs as a virtual network function (VNF) within a network function virtualization (NFV) infrastructure. OBF 234 collects telemetry data from the NF, enabling the use of AI / ML application 236 to optimize and operate the 5G network. OBF application 234 provides near real-time visibility into the performance and operation of the CNF. This collected data is used to optimize the network through a closed-loop automation module (not shown) that performs procedures to provide automated UPF scaling and healing while minimizing manual work and reducing errors.

[0077] In some embodiments, the OBF application 234 performs decoding of the received PM data and publishes the PM data to the data bus 214. In some embodiments, UPF N3 DL PM data is sent to the OBF application 234 every minute. The 5G N3 interface plays the role of transmitting user data from the RAN, such as RAN 112, to the User Plane Function (UPF), enabling the generation of both low and high latency services. In some embodiments, the OBF application 234 performs enrichment of the UPF N3 DL throughput counter in the central inventory 228.

[0078] In some embodiments, the policy manager 204 consumes recommendations from the AI / ML application 236 on an on-demand basis, depending on the network conditions (e.g., predicted) in a network such as the cloud network 108, 5G core network 110, 5G RAN network 112, or 5G transport network 114, via an integrated, high-throughput, low-latency platform for handling real-time data feeds. In some embodiments, the AI / ML application 236 suggests actions based on a learned model, via data on events that the policy manager 204 consumes and triggers actions against the northbound system. In some embodiments, the overall network assurance (day 2) operation is optimized with AI / ML recommendations streamed on a streaming processing platform and consumed by a rule engine such as the policy manager 204.

[0079] In some embodiments, the policy manager 204 triggers an AI / ML recommendation request. In some embodiments, the AI / ML 236 predicts a new threshold and streams the recommendation along with the new PRB count onto the data bus 214. In some embodiments, the AI / ML recommendation is consumed from the data bus 214. In some embodiments, an active policy is triggered for an AI / ML recommendation slice manager action. In some embodiments, the policy manager 204 triggers the slice manager for a PRB configuration change.

[0080] In some embodiments, the policy manager 204 subscribes to incoming events on the data bus 214. In some embodiments, an active policy for scaling in the policy manager 204 is triggered based on the incoming events. In some embodiments, the policy manager 204 calculates the slice utilization percentage and evaluates whether the current threshold, such as a scale-in threshold of 40% or less or a scale-out threshold of 60% or more, has been broken. Depending on whether the threshold has been broken, the policy manager 204 triggers an AI / ML recommendation request. In some embodiments, the AI / ML application 236 predicts the new threshold and streams the recommendation on the data bus 214 along with the new scale-in or scale-out. In some embodiments, the AI / ML recommendation is streamed from the data bus 214 and consumed by the policy manager 204. In some embodiments, a new active policy based on the AI / ML recommendation is generated by the policy manager 204. In some embodiments, in response to an incoming event that triggers a new active policy, the policy manager 204 triggers the slice manager 238 for a UPF SCALEIN or SCALEOUT operation.

[0081] In some embodiments, automated resource optimization with AI predictions across multiple data centers is discussed. In some embodiments, a user request to optimize a slice is received by the slice manager 238 via the UI. In some embodiments, the slice manager 238 streams the request through the data bus 214. In some embodiments, the policy manager 204 consumes the request from the slice manager 238 via the data bus 214 to optimize the slice. In some embodiments, an active policy to optimize the slice is triggered in the policy manager 204. In some embodiments, in response to an incoming event triggering an active policy, the policy manager 204 triggers an AI / ML recommendation request. In some embodiments, the AI / ML application 236 predicts a new threshold and streams the recommendation along with "MIGRATE UPF" onto the data bus 214. In some embodiments, the AI / ML recommendation is streamed via the data bus 214 and consumed by the policy manager 204. In some embodiments, a new active policy is generated in the policy manager 204 based on the AI / ML recommendation. In some embodiments, in response to an incoming event triggering a new active policy, the policy manager 204 triggers the slice manager 238 for a "UPF MIGRATE" operation to the second data center.

[0082] In response to event data received on the data bus 214 lacking fields and / or parameters, events lacking these fields and / or parameters are enriched in the policy correlation and evaluation (PCE) module 226 through an inventory 228 that provides the missing fields and / or parameters, in order to make a decision and take a predetermined action. In some embodiments, this is referred to as inventory enrichment.

[0083] The PCE module 226 logically evaluates and processes events from the data bus 214 based on policies from the policy manager 210. The PCE 226 is configured to identify the root cause of an event, determine the relevant action in accordance with a given policy, and notify the action manager 230 of any relevant action in accordance with the given policy.

[0084] The action manager 230 receives the results after event processing by the PCE 226 and takes corresponding actions regarding those results. In a non-exclusive example, the action manager 230 may send an email, send a request to the API endpoint 232, or take other appropriate actions within the embodiment. The action manager 230 retrieves the status of the actions performed and updates the database 212 so that the user can visualize the job status in the UI 208.

[0085] Figure 3 is a data flow diagram of Method 300 for Physical Resource Block (PRB) Management according to several embodiments.

[0086] Figure 3 is discussed to provide an understanding of the operation of CPEs 100 and 200 through method 300 for PRB management. In some embodiments, method 300 for PRB management is a functional overview of CPEs such as CPE 100 and / or 200. In some embodiments, method 300 for PRB management is performed by processing circuit 602, which is described later with respect to Figure 6. In some embodiments, some or all of the operations of method 300 for PRB management are performed according to instructions corresponding to instructions 606, which are described later with respect to Figure 6.

[0087] Method 300 for PRB management includes operations 302-322, but the operations are not necessarily performed in the order shown. Operations are added, moved, reordered, and / or omitted as appropriate, in accordance with the spirit and scope of the embodiment. In some embodiments, one or more operations of Method 300 for PRB management are repeated. In some embodiments, unless otherwise specified, the operations of Method 300 for PRB management are performed in order.

[0088] In some embodiments, the policy manager 204 analyzes, computes, enriches, and evaluates the collected events. In some embodiments, the user generates a policy template through a user interface (UI), such as UI208 or UI618. The generated policy filters the collected events, enriches the events (e.g., adds any relevant event data), correlates the enriched events, and processes the enriched events for action.

[0089] A user interface (UI), such as UI208 or UI618, is the space in which interaction between humans and machines occurs. The goal of this interaction is to enable effective operation and control of the machine from the human side, while the machine simultaneously provides feedback of information to support the operator's decision-making process. Non-exclusive examples of UIs include computer operating systems, hand tools, heavy machinery operator control, and the interactive aspects of process control. A UI consists of one or more layers, including a human-machine interface (HMI) that interfaces the machine with physical input hardware such as a keyboard, mouse, or gamepad, and output hardware such as a computer monitor, speaker, and printer. A device that implements an HMI is called a human interface device (HID). Other terms for human-machine interface are human-machine interface (MMI), and if the machine in question is a computer, it is a human-computer interface. Additional UI layers may interact with one or more human senses, including haptic UI (touch), visual UI (sight), auditory UI (sound), olfactory UI (smell), vestibular UI (balance), and gustatory UI (taste).

[0090] Actions are triggered based on matching policies. In some embodiments, a CPE core, such as the processing circuit 602 in Figure 6, logically evaluates and processes the collected events. In some embodiments, the CPE core identifies the root cause, determines the relevant action according to a predetermined policy, and instructs an action manager, such as an action manager 230, according to a predetermined network assurance policy. In some embodiments, the action manager 230 collects the results of event processing and takes the respective actions related to the collected results. In non-limiting examples, the action manager 230 may send emails, send requests to application programming interface (API) endpoints, or take other appropriate actions within the embodiment. In some embodiments, the action manager 230 obtains job status feedback to determine the status of executed jobs and updates the backend application in the database so that users can determine the job status through the UI.

[0091] In operation 302 of method 300 for PRB management, one or more vDU324s push PM metrics to OBF234 every minute. In some embodiments, the vDU324 is a gNodeB (gNB) that pushes PM data every minute. The GNodeB is a 3GPP-compliant 5G-NR (new radio) base station implementation. The GNodeB consists of an independent NF that implements the 3GPP-compliant NR RAN protocol.

[0092] Push technology, or server push, is a style of internet-based communication where requests for a given transaction are initiated by a publisher or central server. Push is in contrast to "pull / get," where requests for sending information are initiated by the receiver or client. Push services are often based on pre-expressed information preferences (e.g., subscriptions). Push is called the "publish / subscribe" model. The client "subscribes" to various information "channels" provided by the server, and when new content becomes available on one of those channels, the server pushes that information to the client. The process flows from operation 302 to operation 304.

[0093] In operation 304 of method 300 for PRB management, the OBF application 234 performs decoding of PM data received every minute and publishes the decoded PM data to the data bus 214. The process flows from operation 304 to operation 306.

[0094] In Operation 306 of Method 300 for PRB Management, the Policy Manager 204 subscribes to incoming events on the data bus 214 and event source. The process flows from Operation 306 to Operation 308.

[0095] In Operation 308 of Method 300 for PRB Management, an existing active policy for UE throughput threshold monitoring in Policy Manager 204 is triggered based on one or more incoming events. The process flows from Operation 308 to Operation 310.

[0096] In operation 310 of method 300 for PRB management, the policy manager 204 determines the downlink throughput per UE. As previously stated, the downlink throughput per UE is determined by dividing the guaranteed UE downlink throughput volume by the guaranteed UE downlink throughput time (e.g., dLThptPerUE=GDRB.UEThpVolDL.PCG / GDRB.UEDLThpTime). In some embodiments, the guaranteed UE downlink throughput volume and guaranteed UE downlink throughput time are determined by an SLA. In some embodiments, the SLA establishing the guaranteed UE downlink throughput volume and guaranteed UE downlink throughput time is integrated into a rule-based policy on which the policy manager 204 acts based on event streaming from the data bus 214. The process flows from operation 310 to operation 312.

[0097] In Operation 312 of Method 300 for PRB Management, the Policy Manager 204 determines one or more UE downlink throughput thresholds derived from enrichment of events coming from the Slice Manager 238 and CMaaS rRmpolicy data. In some embodiments, the thresholds include guaranteed throughput, maximum throughput, rRM policy maximum ratio, rRM policy minimum ratio, rRM policy dedicated ratio, low UE downlink throughput, and high UE downlink throughput. In a non-limiting example, the low UE downlink throughput threshold is equal to 120% of the guaranteed throughput. Thus, the provider ensures that the network operates at 120% of the throughput promised to the network subscribers. In another non-limiting example, the high UE downlink throughput is equal to 80% of the maximum throughput. Thus, the provider receives an alert when the UE downlink capacity is at 80% of the maximum, and the provider can take action to ensure that more resources become available before the UE downlink capacity exceeds the maximum. In some embodiments, the network application counter 218 and container counter 220 are enriched with data from the slice manager 238 for throughput data and data from the CMaaS for rRM policy data. The process flows from operation 312 to operation 314.

[0098] In Operation 314 of Method 300 for PRB Management, Policy Manager 204 triggers an AI / ML recommendation request. The process flows from Operation 314 to Operation 316.

[0099] In Operation 316 of Method 300 for PRB Management, AI / ML236 predicts a new throughput threshold and streams the recommendation along with the new PRB count to the data bus 214. The process flows from Operation 316 to Operation 318.

[0100] In Operation 318 of Method 300 for PRB Management, AI / ML recommendations are consumed by the Policy Manager 204 from the Data Bus 214. The process flows from Operation 318 to Operation 320.

[0101] In Operation 320 of Method 300 for PRB Management, a new active policy is generated by Policy Manager 204 based on AI / ML recommendations. The process flows from Operation 320 to Operation 322.

[0102] In Operation 322 of Method 300 for PRB Management, the Policy Manager 204 triggers the Slice Manager 238 for a PRB configuration change based on the triggering of a new active policy.

[0103] Figure 4 is a data flow diagram of Method 400 for automatic level scaling according to several embodiments.

[0104] Figure 4 is discussed to provide an understanding of the operation of CPEs 100 and 200 through method 400 for automatic level scaling. In some embodiments, method 400 for automatic level scaling is a functional overview of CPEs such as CPE 100 and / or 200. In some embodiments, method 400 for automatic level scaling is performed by processing circuit 602, which is described later with respect to Figure 6. In some embodiments, some or all of the operations of method 400 for automatic level scaling are performed according to instructions corresponding to instructions 606, which are described later with respect to Figure 6.

[0105] Method 400 for automatic level scaling includes operations 402-420, but the operations are not necessarily performed in the order shown. Operations are added, moved, reordered, and / or omitted as appropriate, in accordance with the spirit and scope of the embodiment. In some embodiments, one or more operations of Method 400 for automatic level scaling are repeated. In some embodiments, unless otherwise specified, the operations of Method 400 for automatic level scaling are performed in order.

[0106] In operation 402 of method 400 for automatic level scaling, UPF424N3 DL PM data is sent to OBF234 every minute. The process flows from operation 402 to operation 404.

[0107] In operation 404 of method 400 for automatic level scaling, the OBF application 234 enriches the UPF N3 downlink throughput counter data along with the data from the central inventory 228 and sends the enriched UPF N3 downlink throughput counter data to the data bus 214. The process flows from operation 404 to operation 406.

[0108] In operation 406 of method 400 for automatic level scaling, policy manager 204 subscribes to incoming events on data bus 214. The process flows from operation 406 to operation 408.

[0109] In operation 408 of method 400 for automatic level scaling, an existing active policy for scaling is triggered in policy manager 204 based on one or more incoming events. The process flows from operation 408 to 410.

[0110] In operation 410 of method 400 for automatic level scaling, the policy manager 204 determines the slice utilization percentage. In some embodiments, the slice utilization percentage is a quantifier of the efficiency of the slice manager 238 in packing network slices. The process proceeds from operation 410 to operation 412.

[0111] In operation 412 of method 400 for automatic level scaling, policy manager 204 determines whether the slice utilization percentage exceeds the scale-in (40% or less) or scale-out (60% or more) threshold, and an AI / ML recommendation request is triggered depending on whether the scale-in (40% or less) or scale-out (60% or more) threshold has been exceeded.

[0112] In 5G cores and the upcoming 6G cores, UPF (a virtual machine or container implemented as a UPF instance in a cluster) is responsible for sending data to and receiving data from subscribers in PDU sessions. To minimize resource consumption for UPF instances, the number of UPF instances started often depends on the number of PDU sessions, which is controlled by a scaling algorithm. In some embodiments, an AI / ML application 236 makes recommendations for scaling UPF instances packed into containers based on scale-in and scale-out thresholds.

[0113] In operation 414 of method 400 for automatic level scaling, the AI / ML application 236 predicts a new threshold and streams the recommendation along with the new scale-in or scale-out threshold onto the data bus 214. The process flows from operation 414 to operation 416.

[0114] In operation 416 of method 400 for automatic level scaling, AI / ML recommendations are consumed from the data bus 214 by the policy manager 204. The process flows from operation 416 to operation 418.

[0115] In operation 418 of method 400 for automatic level scaling, a new active policy is generated based on AI / ML recommendations. The process flows from operation 418 to operation 420.

[0116] In operation 420 of method 400 for automatic level scaling, in response to triggering a new active policy event, policy manager 204 triggers slice manager 238 for UPF SCALEIN and SCALEOUT operations.

[0117] Figure 5 is a data flow diagram of Method 500 for Data Center Resource Optimization according to several embodiments.

[0118] Figure 5 is discussed to provide an understanding of the operation of CPEs 100 and 200 through Method 500 for Data Center Resource Optimization. In some embodiments, Method 500 for Data Center Resource Optimization is a functional overview of CPEs such as CPE 100 and / or 200. In some embodiments, Method 500 for Data Center Resource Optimization is performed by a processing circuit 602, which is described later with respect to Figure 6. In some embodiments, some or all of the operations of Method 500 for Data Center Resource Optimization are performed according to instructions corresponding to instructions 606, which are described later with respect to Figure 6.

[0119] Method 500 for optimizing data center resources includes operations 502-518, but the operations are not necessarily performed in the order shown. Operations are added, moved, reordered, and / or omitted as appropriate, in accordance with the spirit and scope of the embodiment. In some embodiments, one or more operations of Method 500 for optimizing data center resources are repeated. In some embodiments, unless otherwise specified, the operations of Method 500 for optimizing data center resources are performed in order.

[0120] In operation 502 of method 500 for optimizing data center resources, user 524 requests slice optimization from slice manager 238 via a UI such as UI208 or UI618. The process then proceeds from operation 502 to operation 504.

[0121] In operation 504 of method 500 for optimizing data center resources, the slice manager 238 streams user requests through the data bus 214. The process flows from operation 504 to operation 506.

[0122] In operation 506 of method 500 for optimizing data center resources, the policy manager 204 consumes requests via the data bus 214 to optimize slices from the slice manager 238. The process flows from operation 506 to operation 508.

[0123] In Operation 508 of Method 500 for Data Center Resource Optimization, an existing active policy for slice optimization in Policy Manager 204 is triggered based on one or more incoming events. The process flows from Operation 508 to Operation 510.

[0124] In operation 510 of method 500 for optimizing data center resources, policy manager 204 triggers an AI / ML recommendation request to AI / ML application 236. The process then proceeds from operation 510 to operation 512.

[0125] In operation 512 of method 500 for optimizing data center resources, the AI / ML application 236 predicts new thresholds and streams recommendations along with "MIGRATE UPF" onto the data bus 214. The process flows from operation 512 to operation 514.

[0126] In operation 514 of method 500 for optimizing data center resources, AI / ML recommendations are consumed from the data bus 214 by policy manager 204. The process flows from operation 514 to operation 516.

[0127] In operation 516 of method 500 for optimizing data center resources, a new active policy is generated by policy manager 204 based on AI / ML recommendations. The process flows from operation 516 to operation 518.

[0128] In operation 518 of method 500 for optimizing data center resources, in response to triggering a new active policy, the policy manager 204 triggers the slice manager 238 for a "UPF MIGRATE" operation that migrates UPF from an existing data center to another data center.

[0129] Figure 6 is a high-level functional block diagram of a processor-based system 600 according to several embodiments. In some embodiments, the system 600 is a general-purpose computing device including a hardware processing circuit 602 and a non-temporary computer-readable storage medium 604. The storage medium 604 is encoded with computer instructions 606 (i.e., a set of executable instructions such as the AI ​​Recommendation Automated Assurance Policy Manager) (i.e., it stores computer instructions 606). The execution of instructions 606 by the hardware processing circuit 602 represents a tool (at least in part) that implements some or all of the methods described herein, such as methods 300, 400, 500, etc. (hereinafter referred to as processes and / or methods), according to one or more embodiments.

[0130] The hardware processing circuit 602 is electrically coupled to the computer-readable storage medium 604 via bus 608. The hardware processing circuit 602 is further electrically coupled to the I / O interface 610 via bus 608. The network interface 612 is further electrically connected to the processing circuit 602 via bus 608. The network interface 612 is connected to the network 614, and the processing circuit 602 and the computer-readable storage medium 604 can be connected to external elements via the network 614. The processing circuit 602 is configured to execute computer instructions 606 encoded in the computer-readable storage medium 604 so that the system 600 can be used to perform the aforementioned processes and / or methods such as methods 300, 400, 500 in Figures 3, 4, and 5. In one or more embodiments, the processing circuit 602 is a central processing unit (CPU), a multiprocessor, a distributed processing system, an ASIC (Application Specific Integrated Circuit), and / or a suitable processing unit.

[0131] In one or more embodiments, the computer-readable storage medium 604 is an electronic, magnetic, optical, electromagnetic, infrared, and / or semiconductor system (or apparatus or device). For example, the computer-readable storage medium 604 includes semiconductor or solid-state memory, magnetic tape, removable computer diskette, random access memory (RAM), read-only memory (ROM), rigid magnetic disk, and / or optical disk. In one or more embodiments using an optical disk, the computer-readable storage medium 604 includes compact disk read memory (CD-ROM), compact disk read / write (CD-R / W), and / or digital video disc (DVD).

[0132] In one or more embodiments, the storage medium 604 stores computer instructions 606 configured to be available to the system 600 for performing a portion of the aforementioned process and / or method. In one or more embodiments, the storage medium 604 further stores information such as an AI recommendation automated guarantee policy engine that facilitates the execution of the aforementioned process and / or method.

[0133] System 600 includes an I / O interface 610, such as UI208. The I / O interface 610 is coupled to external circuitry. In one or more embodiments, the I / O interface 610 includes a keyboard, keypad, mouse, trackball, trackpad, touchscreen, cursor directional keys, and / or other suitable I / O interfaces, which are within the assumed scope of this disclosure for transmitting information and commands to processing circuitry 602.

[0134] System 600 further includes a network interface 612 coupled to processing circuit 602. The network interface 612 enables system 600 to communicate with a network 614 to which one or more other computer systems are connected. The network interface 612 includes wireless network interfaces such as BLUETOOTH, WIFI, WiMAX, GPRS, WCDMA, or wired network interfaces such as ETHERNET, USB, IEEE-864. In one or more embodiments, the aforementioned processes and / or methods are implemented in two or more systems 600.

[0135] System 600 is configured to receive information through I / O interface 610. The information received through I / O interface 610 includes one or more instructions, data, and / or other parameters for processing by processing circuit 602. The information is transferred to processing circuit 602 via bus 608. System 600 is configured to receive information about UI, such as UI 618, through I / O interface 610. The information is stored in computer-readable medium 604 as user interface (UI) 208.

[0136] In some embodiments, the aforementioned process and / or method is implemented as a standalone software application for execution by a processing circuit. In some embodiments, the aforementioned process and / or method is implemented as a software application that is part of an additional software application. In some embodiments, the aforementioned process and / or method is implemented as a plug-in to a software application.

[0137] In some embodiments, a process is implemented as a function of a program stored on a non-temporary computer-readable recording medium. Examples of non-temporary computer-readable recording media include, but are not limited to, one or more external / removable and / or internal / built-in storage or memory units, such as optical discs like DVDs, magnetic disks like hard disks, and semiconductor memories like ROMs, RAMs, and memory cards.

[0138] In some embodiments, the system includes a processing circuit and a memory connected to the processing circuit, configured to store executable instructions that facilitate the performance of an operation which, depending on what is performed by the processing circuit, involves a policy manager receiving one or more events from a data bus, triggering an existing assurance policy based on one or more events, triggering a prediction request to an artificial intelligence (AI) / (ML) machine learning application based on the existing assurance policy, receiving change recommendations from the AI / ML application, generating a new assurance policy based on the change recommendations from the AI / ML application, and triggering corrective actions to be taken based on the policy manager triggering the new assurance policy that implements the change recommendations.

[0139] In some embodiments, the execution instructions facilitate the performance of an operation that further includes receiving performance monitor (PM) metrics from a distributed unit (DU) in an observability framework (OBF) application before receiving one or more events from the data bus, depending on the execution by the processing circuit; the OBF application issuing decoded PM metrics to the data bus corresponding to some of the one or more events; and the policy manager subscribing to the data bus to receive one or more events from the data bus.

[0140] In some embodiments, the execution of an executable instruction triggers an existing guarantee policy based on one or more event segments, depending on the execution of the processing circuit, and the policy manager determines the downlink throughput for each user device (UE) based on the guaranteed UE throughput volume over time; the policy manager and based on one or more event segments determine whether the UE downlink throughput has exceeded a low threshold determined by the guaranteed throughput or a high threshold determined by the maximum throughput; and in response to the low or high threshold being exceeded, the policy manager further facilitates the performance of the operation by triggering a prediction request to an AI / ML application.

[0141] In some embodiments, triggering a new assurance policy that implements a change recommendation, and triggering a corrective action to be taken, includes a slice manager modifying one or more eMBB (enhanced mobile broadband) slices based on a request from the policy manager.

[0142] In some embodiments, the execution instruction facilitates the performance of an operation that further includes receiving N3 interface downlink PM metrics from a user plane function (UPF) in an observability framework (OBF) application before receiving one or more events from the data bus, depending on the execution by the processing circuit; enriching the N3 interface downlink PM metrics with data corresponding to the N3 interface downlink PM metrics from a central inventory; and having a policy manager subscribe to the data bus to receive one or more events from the data bus containing the enriched N3 interface downlink PM metrics.

[0143] In some embodiments, the execution instructions are performed by a processing circuit, and the policy manager determines the slice utilization percentage in response to triggering an existing assurance policy based on one or more events which are enriched N3 interface downlink PM metrics; the policy manager determines, based on the enriched N3 interface downlink PM metrics, whether the slice utilization percentage has exceeded a scale-in threshold or a scale-out threshold; and in response to the scale-in threshold or scale-out threshold being exceeded, triggers a prediction request to the AI / ML application to enhance the performance of the operation.

[0144] In some embodiments, the execution instructions facilitate the performance of an operation that further includes receiving slice optimization requests from a user interface (UI) in the slice manager before receiving one or more events from the data bus, depending on whether the execution instructions are performed by the processing circuit; the slice manager streaming the slice optimization requests to the data bus; and the policy manager receiving the slice optimization requests from the data bus.

[0145] In some embodiments, the executable instruction further includes triggering a predictive request to an AI / ML application in response to an existing assurance policy being triggered based on one or more events which are slice optimization requests, in response to execution by a processing circuit; the AI / ML application recommending a "MIGRATE UPF" action based on a slice optimization threshold received with the slice optimization request; and triggering corrective actions to be taken in response to a slice optimization threshold being broken, in response to a new assurance policy implementing the change recommendation, which triggers the execution of a "UPF MIGRATE" operation by the slice manager.

[0146] In some embodiments, the method performed by the processing circuit includes: the policy manager receiving one or more events from a data bus; triggering an existing assurance policy based on one or more events; triggering a prediction request to an artificial intelligence (AI) / (ML) machine learning application based on the existing assurance policy; the policy manager receiving change recommendations from the AI / ML application; the policy manager generating a new assurance policy based on the change recommendations from the AI / ML application; and the policy manager triggering corrective actions to be taken based on the triggering of the new assurance policy that implements the change recommendations.

[0147] In some embodiments, the method further includes: receiving performance monitor (PM) metrics from a distributed unit (DU) in an observability framework (OBF) application before receiving one or more events from the data bus; the OBF application issuing decoded PM metrics to the data bus corresponding to some of the one or more events; and the policy manager subscribing to the data bus in order to receive one or more events from the data bus.

[0148] In some embodiments, the method further includes: the policy manager determining the downlink throughput for each user device (UE) based on guaranteed UE throughput volume over time, in response to triggering an existing guarantee policy based on a portion of one or more events; the policy manager and based on a portion of one or more events determining whether the UE downlink throughput has exceeded a low threshold determined by the guaranteed throughput or a high threshold determined by the maximum throughput; and triggering a prediction request to an AI / ML application in response to the low threshold or high threshold being exceeded.

[0149] In some embodiments, triggering a new assurance policy that implements a change recommendation, and triggering a corrective action to be taken, includes a slice manager modifying one or more eMBB (enhanced mobile broadband) slices based on a request from the policy manager.

[0150] In some embodiments, the method further includes receiving N3 interface downlink PM metrics from a user plane function (UPF) in an Observability Framework (OBF) application before receiving one or more events from the data bus; enriching the N3 interface downlink PM metrics with data corresponding to the N3 interface downlink PM metrics from a central inventory; and having a policy manager subscribe to the data bus to receive one or more events from the data bus containing the enriched N3 interface downlink PM metrics.

[0151] In some embodiments, the method further includes: determining the slice utilization percentage by a policy manager in response to triggering an existing assurance policy based on one or more events which are enriched N3 interface downlink PM metrics; determining whether the slice utilization percentage has exceeded a scale-in threshold or a scale-out threshold by the policy manager and based on the enriched N3 interface downlink PM metrics; and triggering a prediction request to an AI / ML application in response to the scale-in threshold or scale-out threshold being exceeded.

[0152] In some embodiments, the method further includes receiving a slice optimization request from a user interface (UI) in the slice manager before receiving one or more events from the data bus; the slice manager streaming the slice optimization request to the data bus; and the policy manager receiving the slice optimization request from the data bus.

[0153] In some embodiments, the method further includes triggering a prediction request to an AI / ML application in response to triggering an existing assurance policy based on one or more events which are slice optimization requests, and the AI / ML application recommending a “MIGRATE UPF” action based on a slice optimization threshold received with the slice optimization request, and triggering a corrective action to be taken based on triggering a new assurance policy that implements the change recommendation in response to the slice optimization threshold being broken, which in turn triggers the execution of a “UPF MIGRATE” operation by the slice manager.

[0154] In some embodiments, a non-temporary tangible computer-readable storage medium storing a computer program includes instructions to cause a processing circuit to perform operations including: receiving one or more events from a data bus in response to the execution of the computer program; triggering an existing assurance policy based on one or more events; triggering a prediction request to an artificial intelligence (AI) / (ML) machine learning application based on the existing assurance policy; receiving change recommendations from the AI / ML application; generating a new assurance policy based on the change recommendations from the AI / ML application; and triggering corrective actions to be taken based on the new assurance policy implementing the change recommendations.

[0155] In some embodiments, the performance of an operation further includes: receiving performance monitor (PM) metrics from a distributed unit (DU) in an observability framework (OBF) application before receiving one or more events from the data bus, depending on when the instruction is executed by the processing circuit; the OBF application issuing decoded PM metrics to the data bus corresponding to some of the events; and the policy manager subscribing to the data bus to receive one or more events from the data bus.

[0156] In some embodiments, the instructions facilitate the performance of an operation that further includes receiving N3 interface downlink PM metrics from a user plane function (UPF) in an observability framework (OBF) application before receiving one or more events from the data bus, depending on when the instruction is executed by the processing circuit; enriching the N3 interface downlink PM metrics with data corresponding to the N3 interface downlink PM metrics from a central inventory; and having a policy manager subscribe to the data bus to receive one or more events from the data bus containing the enriched N3 interface downlink PM metrics.

[0157] In some embodiments, the performance of an operation is enhanced by the following: receiving a slice optimization request from a user interface (UI) in the slice manager before receiving one or more events from the data bus, depending on when the instruction is executed by the processing circuit; the slice manager streaming the slice optimization request to the data bus; and the policy manager receiving the slice optimization request from the data bus.

[0158] The above outlines some of the features of the embodiments so that those skilled in the art may better understand them. Those skilled in the art will understand that the embodiments can be readily used as a basis for designing or modifying other processes and structures to achieve the same objectives and / or advantages as the embodiments presented herein. Those skilled in the art will further recognize that such equivalent configurations will not depart from the spirit and scope of the embodiments, and that various changes, substitutions, and modifications will not depart from the spirit and scope of the embodiments.

Claims

1. Processing circuit and A memory connected to the processing circuit, which, in accordance with what is executed by the processing circuit, The policy manager receives one or more events from the data bus, Based on one or more of the aforementioned events, trigger an existing warranty policy, Based on the aforementioned existing assurance policy, the following will be triggered: The policy manager receives change recommendations from the AI / ML application, The policy manager generates a new assurance policy based on the change recommendations from the AI / ML application, The policy manager triggers corrective actions to be taken based on the triggering of the new assurance policy that implements the change recommendations in response to one or more events received from the data bus after the new assurance policy has been generated. Memory configured to store executable instructions that facilitate the performance of operations including, A system equipped with these features.

2. The executable instruction is executed by the processing circuit, Before receiving one or more events from the data bus, the Observability Framework (OBF) application receives performance monitor (PM) metrics from the distributed unit (DU), The OBF application issues decoded PM metrics corresponding to some of the events to the data bus, The policy manager subscribes to the data bus in order to receive the one or more events from the data bus, The system according to claim 1, which further enhances the performance of operations including the following.

3. The executable instruction is executed by the processing circuit, In response to triggering the existing guarantee policy based on one or more of the aforementioned events, the policy manager determines the downlink throughput for each user equipment (UE) based on the guaranteed UE throughput volume over time, The policy manager and based on the portion of the one or more events determine whether the UE downlink throughput has exceeded a low threshold determined by guaranteed throughput or a high threshold determined by maximum throughput. In response to the breach of the low threshold or the high threshold, the prediction request to the AI / ML application is triggered. The system according to claim 2, which further enhances the performance of operations including the following.

4. The system according to claim 3, wherein triggering the corrective action to be taken, based on the triggering of the new assurance policy that implements the recommended change, comprises a slice manager modifying one or more eMBB (enhanced mobile broadband) slices based on a request from the policy manager.

5. The executable instruction is executed by the processing circuit, Before receiving one or more events from the data bus, the Observability Framework (OBF) application receives N3 interface downlink PM metrics from the User Plane Function (UPF), The aforementioned N3 interface downlink PM metrics are enriched with data corresponding to the aforementioned N3 interface downlink PM metrics from the central inventory, The policy manager subscribes to the data bus to receive one or more events from the data bus that include the enriched N3 interface downlink PM metrics, The system according to claim 1, which further enhances the performance of operations including the following.

6. The executable instruction is executed by the processing circuit, In response to triggering the existing assurance policy based on one or more events which are the enriched N3 interface downlink PM metrics, the policy manager determines the slice utilization percentage, Based on the policy manager and the enriched N3 interface downlink PM metrics, it is determined whether the slice utilization percentage has exceeded the scale-in threshold or scale-out threshold. In response to the breach of the scale-in threshold or the scale-out threshold, the prediction request to the AI / ML application is triggered. The system according to claim 5, which further enhances the performance of operations including the following.

7. The executable instruction is executed by the processing circuit, Before receiving one or more events from the data bus, the slice manager receives a slice optimization request from the user interface (UI), The slice manager streams the slice optimization request to the data bus, The policy manager receives the slice optimization request from the data bus, The system according to claim 1, which further enhances the performance of operations including the following.

8. The executable instruction is executed by the processing circuit, In response to triggering the existing guarantee policy based on the one or more events which are slice optimization requests, the prediction request to the AI / ML application is triggered. The AI / ML application recommends the "MIGRATE UPF" action based on the slice optimization threshold received along with the slice optimization request, To further enhance the performance of operations, including The system according to claim 7, wherein triggering the corrective action to be taken, based on triggering the new assurance policy that implements the change recommendation in response to the slice optimization threshold being violated, includes triggering the slice manager to perform a "UPF MIGRATE" operation.

9. A method executed by a processing circuit, The policy manager receives one or more events from the data bus, Based on one or more of the aforementioned events, trigger an existing warranty policy, Based on the aforementioned existing assurance policy, the following will be triggered: The policy manager receives change recommendations from the AI / ML application, The policy manager generates a new assurance policy based on the change recommendations from the AI / ML application, The policy manager triggers corrective actions to be taken based on the triggering of the new assurance policy that implements the change recommendations in response to one or more events received from the data bus after the new assurance policy has been generated. A method for providing this.

10. Before receiving one or more events from the data bus, the Observability Framework (OBF) application receives performance monitor (PM) metrics from the distributed unit (DU), The OBF application issues decoded PM metrics corresponding to some of the events to the data bus, The policy manager subscribes to the data bus in order to receive the one or more events from the data bus, The method according to claim 9, further comprising:

11. In response to triggering the existing guarantee policy based on one or more of the aforementioned events, the policy manager determines the downlink throughput for each user equipment (UE) based on the guaranteed UE throughput volume over time, The policy manager and based on the portion of the one or more events determine whether the UE downlink throughput has exceeded a low threshold determined by guaranteed throughput or a high threshold determined by maximum throughput. In response to the breach of the low threshold or the high threshold, the prediction request to the AI / ML application is triggered. The method according to claim 10, further comprising:

12. The method according to claim 11, wherein triggering the corrective action to be taken, based on triggering the new assurance policy that implements the recommended change, comprises modifying one or more eMBB (enhanced mobile broadband) slices by a slice manager based on a request from the policy manager.

13. Before receiving one or more events from the data bus, the Observability Framework (OBF) application receives N3 interface downlink PM metrics from the User Plane Function (UPF), The aforementioned N3 interface downlink PM metrics are enriched with data corresponding to the aforementioned N3 interface downlink PM metrics from the central inventory, The policy manager subscribes to the data bus to receive one or more events from the data bus that include the enriched N3 interface downlink PM metrics, The method according to claim 9, further comprising:

14. In response to triggering the existing assurance policy based on one or more events which are the enriched N3 interface downlink PM metrics, the policy manager determines the slice utilization percentage, Based on the policy manager and the enriched N3 interface downlink PM metrics, it is determined whether the slice utilization percentage has exceeded the scale-in threshold or scale-out threshold. In response to the breach of the scale-in threshold or the scale-out threshold, the prediction request to the AI / ML application is triggered. The method according to claim 13, further comprising:

15. Before receiving one or more events from the data bus, the slice manager receives a slice optimization request from the user interface (UI), The slice manager streams the slice optimization request to the data bus, The policy manager receives the slice optimization request from the data bus, The method according to claim 9, further comprising:

16. In response to triggering the existing guarantee policy based on the one or more events which are slice optimization requests, the prediction request to the AI / ML application is triggered. The AI / ML application recommends the "MIGRATE UPF" action based on the slice optimization threshold received along with the slice optimization request, Furthermore, The method of claim 15, wherein triggering the corrective action to be taken, based on triggering the new assurance policy that implements the change recommendation in response to the slice optimization threshold being violated, includes triggering the slice manager to perform a "UPF MIGRATE" operation.

17. A non-temporary, tangible, computer-readable storage medium that stores computer programs, The computer program, in accordance with its execution, The policy manager receives one or more events from the data bus, Based on one or more of the aforementioned events, trigger an existing warranty policy, Based on the aforementioned existing assurance policy, the following will be triggered: The policy manager receives change recommendations from the AI / ML application, The policy manager generates a new assurance policy based on the change recommendations from the AI / ML application, The policy manager triggers corrective actions to be taken based on the triggering of the new assurance policy that implements the change recommendations in response to one or more events received from the data bus after the new assurance policy has been generated. A non-temporary, tangible, computer-readable storage medium containing instructions that cause a processing circuit to execute an operation including [a specific operation].

18. The instruction is executed by the processing circuit, Before receiving one or more events from the data bus, the Observability Framework (OBF) application receives performance monitor (PM) metrics from the distributed unit (DU), The OBF application issues decoded PM metrics corresponding to some of the events to the data bus, The policy manager subscribes to the data bus in order to receive the one or more events from the data bus, A non-temporary tangible computer-readable storage medium according to claim 17, which further enhances the performance of operations including the following:

19. The instruction is executed by the processing circuit, Before receiving one or more events from the data bus, the Observability Framework (OBF) application receives N3 interface downlink PM metrics from the User Plane Function (UPF), The aforementioned N3 interface downlink PM metrics are enriched with data corresponding to the aforementioned N3 interface downlink PM metrics from the central inventory, The policy manager subscribes to the data bus to receive one or more events from the data bus that include the enriched N3 interface downlink PM metrics, A non-temporary tangible computer-readable storage medium according to claim 17, which further enhances the performance of operations including the following:

20. The instruction is executed by the processing circuit, Before receiving one or more events from the data bus, the slice manager receives a slice optimization request from the user interface (UI), The slice manager streams the slice optimization request to the data bus, The policy manager receives the slice optimization request from the data bus, A non-temporary tangible computer-readable storage medium according to claim 17, which further enhances the performance of operations including the following:

Images