Computing-network operation administration and maintenance system and method, and storage medium

Abstract

Provided in the present application are a computing-network operation administration and maintenance system and method, and a storage medium. The system comprises: a network sensing component, which is used for network information sensing and diffusion of network nodes and links; a computing power sensing component, which is used for interfacing with computing power information collected by a cloud-side computing power collection component deployed in a cloud resource pool and receiving computing power information of a computing power gateway device and diffusing same; a computing power routing computation component, which separately interfaces with the network sensing component and the computing power sensing component, and is used for summarizing metric information respectively corresponding to a network and a computing power and dynamically determining computing power routing on the basis of the metric information; a computing power routing forwarding component, which is used for receiving the computing power routing determined by the computing power routing computation component and providing computing-network integrated forwarding for service messages on the basis of the computing power routing; and a computing-network operation administration and maintenance component, which executes operation administration and maintenance testing on the basis of received objects from the computing power sensing component and the computing power routing computation component and reports a testing result. Thus, computing-network services can be dynamically adjusted, thereby improving the service quality.

Description

Computing network operation and maintenance management system, method and storage medium Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a computing network operation and maintenance management system, method, and storage medium. Background Art

[0002] The future development goal of computing power networks is to achieve computing-network integration. Computing-network integration has two major driving forces. The first is to leverage the inherent distributed characteristics of network facilities to help computing power resources be distributedly deployed to meet the diverse needs of various applications for latency, energy consumption, and security; the second is to achieve coordinated scheduling of computing power and networks to meet the integrated needs of computing resources and network connections in different business scenarios.

[0003] At present, computing networks usually use network equipment to sense the status of computing resources, and combine the network resource status to perform preferential processing, select the best computing instance and the path to the computing instance. In order to achieve better service guarantee, on the one hand, it is necessary to combine computing network calculations based on the real-time status of the network and computing resources as much as possible. On the other hand, it is necessary to detect whether the computing network service quality meets the standards as quickly as possible to trigger dynamic calculation and adjustment of computing network services. The existing network-side operation, administration and maintenance (OAM) is very mature, but these OAMs cannot directly cover computing instance nodes. At the same time, there is no unified OAM standard on the computing side. The computing network is network-centric. How to implement computing network services and ensure service quality is an urgent problem that needs to be solved.

[0004] Summary of the Invention

[0005] This application aims to provide a computing network operation and maintenance management system, method and storage medium to realize the joint calculation of network and computing resources, calculate and adjust the computing network business dynamics, improve the computing network service quality and enhance the user experience.

[0006] The embodiment of the present application provides a computing network operation and maintenance management system, which is applied to a computing power gateway device, and the system includes: a network perception component, a computing power perception component, a computing power routing calculation component, a computing power routing forwarding component, and a computing network operation and maintenance management component;

[0007] The network awareness component is used for awareness and diffusion of network information of network nodes and links;

[0008] The computing power perception component is used to connect to the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and to receive and disseminate the computing power information of the computing power gateway device;

[0009] The computing power routing calculation component is connected to the network perception component and the computing power perception component respectively, and is used to summarize the measurement indicator information corresponding to the network and computing power respectively, and dynamically determine the computing power routing according to the measurement indicator information;

[0010] The computing power routing forwarding component is used to receive the computing power routing determined by the computing power routing calculation component, and provide computing network integrated forwarding for the service message according to the computing power routing;

[0011] The computing network operation and maintenance management component performs operation and maintenance management detection based on the objects received from the computing power perception component and the computing power routing calculation component and reports the detection results.

[0012] The present application also provides a method for network operation and maintenance management, wherein the method is applied to a network operation and maintenance management system, and the method includes:

[0013] The computing power perception component obtains computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and sends enabling information to the computing network operation and maintenance management component based on the computing power information to start the computing network operation and maintenance management component;

[0014] The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0015] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, so that the computing instance generates an operation and maintenance management message response according to the deployed operation and maintenance management service and the operation and maintenance management message protocol;

[0016] The computing power routing and forwarding component receives the operation and maintenance management message response of the computing instance, and sends the operation and maintenance management message response to the computing network operation and maintenance management component;

[0017] The computing network operation and maintenance management component generates an operation and maintenance management detection result according to the operation and maintenance management message response, and determines an operation and maintenance management detection notification message according to the detection object information;

[0018] The computing power perception component receives the operation maintenance management detection notification message and spreads the computing power information corresponding to the operation maintenance management detection notification message;

[0019] The upstream node receives and processes the computing power information, recalculates the computing power routing and sends it to the computing power routing forwarding component.

[0020] The present application also provides a method for network operation and maintenance management, wherein the method is applied to a network operation and maintenance management system, and the method includes:

[0021] The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component;

[0022] The computing power routing calculation component generates a computing power routing table according to the measurement process information, and sends the computing power routing table to the computing power routing forwarding component;

[0023] The computing power routing calculation component sends enabling information to the computing network operation and maintenance management component according to the computing power routing table;

[0024] The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0025] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component;

[0026] The computing network operation and maintenance management component generates a detection result of the computing power routing table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component;

[0027] The computing power routing calculation component determines a service level agreement target gap based on the detection result, recalculates a computing power routing table based on the service level agreement target gap, and sends the computing power routing table to the computing power routing forwarding component.

[0028] The present application also provides a method for network operation and maintenance management, wherein the method is applied to a network operation and maintenance management system, and the method includes:

[0029] The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component;

[0030] The computing power routing calculation component generates a computing power routing table according to the measurement process information, and sends the computing power routing table to the computing power routing forwarding component;

[0031] The computing power routing forwarding component determines a flow affinity table based on the received computing power routing table and the computing power service access request of the terminal user, and sends enabling information to the computing network operation and maintenance management component;

[0032] The computing network operation and maintenance management component constructs an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0033] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component;

[0034] The computing network operation and maintenance management component determines a detection result based on the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component;

[0035] The computing power routing calculation component determines the service level agreement target gap corresponding to the detection result, calculates the path compensation according to the service level agreement target gap, and sends the path compensation to the computing power routing forwarding component to refresh the flow affinity table.

[0036] An embodiment of the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs are executed by one or more processors to implement any of the computing network operation, maintenance and management methods described in the embodiments of the present application.

[0037] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0039] Figure 1 is a schematic diagram of the working mechanism of a computing power network;

[0040] FIG2 is a schematic diagram of the structure of a computing network operation and maintenance management system provided in an embodiment of the present application;

[0041] FIG3 is an example diagram of the operation of a computing network operation and maintenance management system provided in an embodiment of the present application;

[0042] FIG4 is an example diagram of an OAM architecture provided in an embodiment of the present application;

[0043] FIG5 is a flowchart of a method for operation, maintenance and management of a computing network provided in an embodiment of the present application;

[0044] FIG6 is an example diagram of a method for operation, maintenance and management of a computing network provided in an embodiment of the present application;

[0045] 7 is a flowchart of another method for network operation, maintenance and management provided by an embodiment of the present application;

[0046] FIG8 is an example diagram of a method for operation, maintenance and management of a computing network provided in an embodiment of the present application;

[0047] 9 is a flowchart of another method for network operation, maintenance and management provided by an embodiment of the present application;

[0048] FIG10 is an example diagram of a method for network operation, maintenance and management provided in an embodiment of the present application;

[0049] FIG11 is a schematic structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] It should be understood that the specific implementations described herein are only used to explain the present application and are not used to limit the present application.

[0051] In the subsequent description, suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of this application and have no specific meaning. Therefore, "module", "component" or "unit" can be used interchangeably.

[0052] As shown in Figure 1, computing network technology generally uses ServiceID (for example, ANYCAST address or pure numbers) as the service identifier for computing services to initiate access services. In order to meet the joint scheduling of computing networks, computing networks require dual perception of both network and computing power. Traditional network OAM, as part of network perception technology, mainly implements fault management (FM) and performance monitoring (PM). FM can include fault management such as connectivity and path, while PM can include performance monitoring such as packet loss, latency, and bandwidth. Traditional OAM can cover FM and PM of network domain links and bearer services, but has not yet achieved FM perception and PM perception from the network to computing services.

[0053] In order to achieve computing power perception, a computing power collection module is usually deployed on the cloud side to obtain information such as processing delay, processing bandwidth, central processing unit (CPU) occupancy, memory occupancy, and continuous number of computing power resources, and send it to the network domain egress gateway through the notification protocol. The egress gateway then notifies the ingress gateway of the network computing power.

[0054] To achieve network awareness, the network domain uses technologies such as LINK OAM to detect link metrics as part of link attributes. It uses the Interior Gateway Protocol (IGP) to propagate links and link attributes, as well as achieve head node network awareness. This information is synchronized to the controller via the Border Gateway Protocol-Link State (BGP-LS). Furthermore, network nodes directly send telemetry information, such as link occupancy and traffic, to the controller. Joint computing and network calculations are performed through the ingress gateway or intelligent controller, ultimately forming a computing power routing table that guides computing network traffic.

[0055] When a terminal's message carrying the ServiceID reaches the ingress gateway for the first time, a flow affinity table is generated to ensure that the current session instance remains unchanged until the session ends. Although the above technology implements basic computing network functions, existing solutions still have the following problems:

[0056] 1. The egress gateway connects to the computing instance through the Data Communication Network (DCN). However, the DCN network may experience performance degradation or disconnection. At the same time, as the access load increases, the performance indicators of the computing instance will also deteriorate. Although instance indicators can be obtained through computing power collection and measurement and notified to the egress gateway through the perception protocol, this cannot cover the indicators of the DCN network segment. To avoid the high-frequency reporting of computing power putting pressure on the control plane, slow cycle reporting is usually adopted. However, once the DCN network or computing instance fails or degrades, computing power rerouting convergence cannot be triggered in time to avoid routing black holes.

[0057] 2. When calculating computing power routing, it is usually necessary to meet the requirements of the computing network joint service-level agreement (SLA). However, the computing power routing table for a specific service ID on the ingress gateway cannot evaluate and prove whether the computing network joint computing target meets the expected requirements of the computing network SLA. Therefore, it is necessary to trigger the control plane calculation in real time through the OAM detection mechanism to generate new table entries to achieve a dynamic closed loop of service capabilities at the service ID level. However, existing telecommunications-grade OAM technology cannot directly implement detection at the computing power routing level.

[0058] 3. After completing a connection session accessing the computing service, packets are forwarded to the corresponding compute instance by matching the flow affinity table. As access volume increases, the processing latency of the compute instance may deteriorate, impacting the user experience of the completed connection. However, existing carrier-grade network OAM technologies cannot directly perform end-to-end detection of flows matching the flow affinity table in the computing network. This allows for dynamic evaluation of degradation indicators, path adjustments, and compensation to ensure a consistent experience for end users.

[0059] In response to the above problems, the embodiments of the present application provide a computing network operation and maintenance management system, method and storage medium to achieve rapid convergence of computing power routing and ensure the quality of computing network services, which has strong feasibility and scalability.

[0060] FIG2 is a schematic structural diagram of a computing network operation and maintenance management system provided in an embodiment of the present application. The present application is applicable to computing network operation and maintenance management, and is generally applied to computing power gateway devices. The system can be implemented in software and / or hardware. Referring to FIG2 , the system provided in an embodiment of the present application specifically includes: a network perception component 110, a computing power perception component 120, a computing power routing calculation component 130, a computing power routing forwarding component 140, and a computing network operation and maintenance management component 150;

[0061] The network perception component 110 is used to perceive and disseminate network information of network nodes and links; the computing power perception component 120 is used to connect to the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and to receive and disseminate the computing power information of the computing power gateway device; the computing power routing calculation component 130 is connected to the network perception component 110 and the computing power perception component 120 respectively, and is used to summarize the measurement indicator information corresponding to the network and computing power, and dynamically determine the computing power routing according to the measurement indicator information; the computing power routing forwarding component 140 is used to receive the computing power routing determined by the computing power routing calculation component 130, and provide computing network integrated forwarding for business messages according to the computing power routing; the computing network operation and maintenance management component 150 performs operation and maintenance management detection based on the objects received from the computing power perception component 120 and the computing power routing calculation component 130 and reports the detection results.

[0062] The network perception component 110 can be a device for sensing the network. The network perception component can be deployed on the control plane. The network perception component 110 can be deployed in entry gateways, border gateways, network telemetry devices, etc. The network perception component 110 can perceive the quality of network nodes and links, and spread the perceived network information. It can be understood that the information perceived by the network perception component 110 may include but is not limited to latency, bandwidth, bandwidth occupancy, traffic information, etc.

[0063] The computing power perception component 120 can be deployed on the control plane, can be connected to the cloud-side computing power collection component deployed in the cloud-side resource pool, and can perceive the indicators of the computing instances collected by the cloud-side computing power collection component. In addition, the computing power perception component 120 can also be deployed on the export computing power gateway device or network controller, can receive computing power information and spread it. Since the computing power status is highly time-varying, the notification interval is usually kept at the same level as the routing protocol, which can reduce the pressure on the control plane.

[0064] The computing power routing calculation component 130 can be deployed on the control plane and can connect to the network perception component 110 and the computing power perception component 120. It can receive the metric information sensed by the network perception component 110 and the computing power perception component 120 respectively, and this metric information can be used to determine the computing power routing. The computing power routing calculation component 130 is also connected to the routing forwarding component 140, and the computing power routing calculation component 130 can send the determined computing power routing to the computing power routing forwarding component 140.

[0065] The computing power routing forwarding component 140 can receive the computing power routing issued, and provide computing network integrated forwarding services for the business messages carrying the ServiceID, and can provide forwarding services for both computing power and business messages within the network.

[0066] The computing network operation and maintenance management component 150 can receive objects provided by the computing power perception component and the computing power routing computing component, perform operation and maintenance management detection on the objects, and report the detection results. The operation and maintenance management detection that can be implemented by the computing network operation and maintenance management component 150 includes but is not limited to connectivity fault detection, packet loss rate detection, delay detection, etc. The objects of detection may include one or more of the flows corresponding to the computing instance, computing power routing table, and flow affinity table.

[0067] The embodiments of the present application provide a system that achieves rapid convergence of computing and routing and reliable quality assurance of computing network services. The slow lane control plane is composed of a network perception component, a computing power perception component, a computing power routing calculation component, and a computing power routing forwarding component. The slow lane control plane implements service forwarding planning for collection, measurement, and calculation, and completes basic service action control through rule issuance. The computing network operation and maintenance management component and the computing power routing forwarding component implement a fast lane, realize OAM rapid collection, measurement, and calculation correction, and perform service optimization through control, so that the computing network system becomes a negative feedback closed-loop system, providing computing network services with a better consistency experience.

[0068] Based on the above application embodiment, the computing power perception component is used to connect to the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and to notify the computing power information to the protocol diffusion, including:

[0069] The computing power perception component 120 is deployed on the export computing power gateway device or network controller to receive the computing power information collected by the cloud-side computing power collection component and notify the computing power information to the protocol diffusion.

[0070] In an embodiment of the present application, the computing power perception component 120 can have multiple deployment locations within the OAM system. When the computing power perception component 120 is deployed on the control plane, for example, an export computing power gateway device or a network controller, it can be connected to the cloud-side computing power collection component and obtain the computing power information collected by the cloud-side computing power collection component. The computing power perception component 120 can also notify the computing power information to the general protocol to spread the computing power information. Among them, the cloud-side computing power collection component can be a collection device for computing instance indicators deployed in the cloud resource pool. The cloud resource pool can realize the resource pool by virtualizing one or more physical hosts. Computing power computing instances can be deployed in the cloud resource pool. The cloud-side computing power collection component can collect indicators of one or more computing power computing instances deployed in the cloud resource pool. The indicators may include but are not limited to processing delay, CPU occupancy, memory occupancy, bandwidth occupancy, etc.

[0071] In some application embodiments, the system collects computing network status based on the cloud-side computing power collection component, the computing power perception component 120, and the network perception component 111. The computing network status collection includes at least:

[0072] The cloud-side computing power collection component notifies the computing power perception component 120 of the collected computing power information to notify the protocol to perform network domain diffusion to form a computing power status database;

[0073] The network domain collects network status information through the network awareness component 110 to form a network status database.

[0074] In the embodiment of the present application, computing network status collection can be implemented by the computing power perception component 120 and the network perception component 110. The computing power perception component 120 can receive computing power information passed by the cloud-side computing power collection component and spread the computing power information through the protocol to generate a computing power status database. It is understandable that the computing power status database can store computing power information collected by the cloud-side computing power collection component connected to the computing power perception component 120. The network perception component 110 can collect network status information in the network domain, and the network status information can be stored in the network status database.

[0075] In some other application embodiments, the system performs computing network joint detection and analysis based on the computing power routing calculation component 130 and the computing power routing forwarding component 140. The computing network joint detection and analysis includes at least one of the following:

[0076] The computing power routing calculation component 130 determines a computing power routing table based on the computing power measurement index information and the network measurement index information, and sends the computing power routing table to the computing power routing forwarding component 140. In addition, the computing power routing calculation component 130 sends enabling information to the computing network operation and maintenance management component 150 based on the computing power routing table to trigger the computing network operation and maintenance management component 150 to perform detection and feedback the detection results;

[0077] The computing power routing and forwarding component 140 generates a flow affinity table based on the computing power routing table and the computing power service access request of the terminal user, and the computing power routing and forwarding component 140 sends an enabling message to the computing network operation and maintenance management component 150 to trigger the computing network operation and maintenance management component 150 to perform detection and feedback the detection results;

[0078] The computing power routing calculation component 130 performs a service level agreement target gap analysis based on the detection results.

[0079] In an embodiment of the present application, the computing network joint detection and analysis process can be implemented by the computing power routing calculation component 130 and the computing power routing forwarding component 140. The computing network joint calculation process can include the computing power routing calculation component 130 determining the computing power routing table according to the computing power measurement index information and the network measurement index information. The computing power routing calculation component 130 can send the calculated computing power routing table to the computing power routing forwarding component 140. At the same time, when generating the computing power routing table, the computing power routing calculation component 130 can send enabling information to the computing network operation and maintenance management component 150. It can be understood that the enabling information can be information that triggers the computing network operation and maintenance management component 150 to detect the generated computing power routing table, and enables the computing network operation and maintenance management component 150 to report the detection results generated by the detection. The computing network joint detection and analysis process can also include the computing power routing forwarding component 140 generating a flow affinity table through the computing power routing table and the computing power service access request of the terminal user. The computing power routing forwarding component 140 can send enabling information to the computing network operation and maintenance management component 150, so that the computing network operation and maintenance management component 150 can detect the flow corresponding to the flow affinity table according to the enabling information, thereby generating and feeding back the detection results. In addition, the computing power routing calculation component 130 can also perform a service level agreement target gap analysis based on the detection results generated by the computing network operation and maintenance management component 150 to determine the gap between the current detection results and the service level agreement target. It can be understood that the service level agreement target gap analysis can be determined by comparing the indicators of the detection results. For example, the service level agreement target gap analysis can include but is not limited to analyzing the gap in computing network packet loss, analyzing the gap in computing network latency, and analyzing the gap in other computing network indicators.

[0080] Based on the above application embodiment, it also includes: the computing power routing calculation component 130 receives the flow affinity table and manages and maintains the user session according to the flow affinity table.

[0081] In an embodiment of the present application, the computing power routing calculation component 130 can also obtain the flow affinity table generated by the computing power routing forwarding component 140, and the computing power routing calculation component 130 can also manage and maintain user sessions based on the flow affinity table.

[0082] In some application embodiments, it also includes: the system measures the detection object according to the computing network operation and maintenance management component and the computing power routing and forwarding component, wherein the detection object includes: at least one of the computing instance, the computing power routing table, and the flow affinity table corresponding to the flow.

[0083] Based on the above application embodiment, the detection object includes a computing instance, and the system measures the detection object according to the computing network operation and maintenance management component and the computing power routing and forwarding component, including:

[0084] The computing network operation and maintenance management component receives the enabling information sent by the computing power perception component based on the computing power information, assembles the operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0085] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component;

[0086] The computing network operation and maintenance management component detects the computing instance based on the operation and maintenance management message response, and feeds back the detection results to the computing power routing calculation component.

[0087] In an embodiment of the present application, the computing instance can be detected by the computing network operation and maintenance management component 150 and the computing power routing and forwarding component 140. The detection process may include: the computing network operation and maintenance management component can receive the enabling information sent by the computing power perception component 120 based on the computing power information it is connected to. When the computing network operation and maintenance management component 150 receives the enabling information, it can assemble an operation and maintenance management message according to the enabling information. It can be understood that the message protocol format of the operation and maintenance management message assembled by the computing network operation and maintenance management component 150 can be determined by its pre-configured protocol format. The computing network operation and maintenance management component 150 can send the encapsulated operation and maintenance management message to the computing power routing and forwarding component 140. The computing power routing and forwarding component 140 can forward the operation and maintenance management message to the computing instance. The computing instance can be deployed in the cloud-side resource pool, and the computing power routing and forwarding component 140 can be deployed in the ingress gateway and the egress gateway respectively, and can forward the operation and maintenance management message from the intranet to the extranet. The computing instance can respond to the received operation and maintenance management message to generate an operation and maintenance management message reply. The computing power routing forwarding component 140 can receive the operation and maintenance management message reply generated by the computing instance and forward the operation and maintenance management message reply to the computing network operation and maintenance management component 150. The computing network operation and maintenance management component 150 can detect the computing instance based on the received operation and maintenance management message reply, and can determine the packet loss rate, computing network delay and other indicators corresponding to the computing instance as the detection result. The computing network operation and maintenance management component 150 can feed back the generated detection result to the computing power routing computing component. In this feedback process, the computing network operation and maintenance management component 150 can feed back the detection result to the computing power perception component 120, and the computing power perception component 120 will notify the protocol of the detection result for dissemination, and the computing power routing computing component 130 can receive the detection result. It is understandable that the network operation and maintenance management component 150 can also compare the detection result with the preset threshold before feedback. When it is determined that the detection result exceeds the preset threshold, the detection result can be fed back. If it does not exceed the preset threshold, the detection result may not be fed back.

[0088] Based on the above application embodiment, the detection object includes a computing power routing table. The system measures the detection object according to the computing network operation and maintenance management component 150 and the computing power routing forwarding component 140, including:

[0089] The computing power routing forwarding component 140 receives the computing power routing table determined by the computing power routing calculation component 130 according to the computing power measurement index information and the network measurement index information; the computing network operation and maintenance management component 150 receives the enabling information issued by the computing power routing calculation component 130 according to the computing power routing entry of the computing power routing table, assembles the operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing forwarding component 140; the computing power routing forwarding component 140 deployed at the inlet gateway encapsulates and forwards the operation and maintenance management message to the computing power routing forwarding component 140 deployed at the egress gateway; the computing power routing forwarding component 140 deployed at the egress gateway encapsulates and forwards the operation and maintenance management message to the computing power routing forwarding component 140 deployed at the egress gateway; The sending component 140 decapsulates the operation and maintenance management message, sends the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, encapsulates the operation and maintenance management message response and forwards it to the computing power routing forwarding component 140 deployed at the entry gateway; the computing power routing forwarding component 140 deployed at the entry gateway decapsulates the operation and maintenance management message response, and sends the operation and maintenance management message response to the computing network operation and maintenance management component 150; the computing network operation and maintenance management component 150 detects the computing power routing table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing computing component 130.

[0090] In an embodiment of the present application, the computing power routing calculation component 130 can determine the computing power routing table through the computing power measurement index information and the network measurement index information, and send the computing power routing table to the computing power routing forwarding component 140. At the same time, the computing power routing calculation component 130 can determine the enabling information according to the computing power routing entries in the computing power routing table. The enabling information can trigger the computing network operation and maintenance management component 150 to detect the computing power routing table. The computing power routing calculation component 130 can send the generated enabling information to the computing network operation and maintenance management component 150. The computing network operation and maintenance management component 150 can assemble the computing network operation and maintenance management message according to the enabling information, and send the encapsulated computing network operation and maintenance management message to the computing power routing forwarding component 140 deployed at the inlet gateway. The computing power routing forwarding component 140 deployed at the inlet gateway encapsulates the computing network operation and maintenance management message and forwards the encapsulated operation and maintenance management message to the computing power routing forwarding component 140 deployed at the egress gateway. The computing power routing forwarding component 140 deployed at the egress gateway can decapsulate the operation and maintenance management message and send the operation and maintenance management message to the computing instance to obtain the operation and maintenance management message of the computing instance in response to the maintenance operation management message. The computing power routing forwarding component 140 deployed at the egress gateway can encapsulate the operation and maintenance management message response and transmit it to the computing power routing forwarding component 140 deployed at the ingress gateway. The computing power routing forwarding component 140 deployed at the ingress gateway decapsulates the operation and maintenance management message response and transmits the operation and maintenance management message response to the computing network operation and maintenance management component 150. It can be understood that the operation and maintenance management message and the operation and maintenance management message response can be encapsulated and decapsulated by the computing power routing forwarding component 140 deployed at the egress gateway and the ingress gateway, thereby achieving full-level message transmission and unified management of OAM detection messages. The computing network operation and maintenance management component 150 can detect the computing power routing table based on the operation and maintenance management message response and determine the performance indicator of the path corresponding to the computing power routing table entry as the detection result. The computing network operation and maintenance management component 150 can also provide feedback on the determined detection result to facilitate subsequent adjustment and correction of the computing power routing table result.

[0091] In other application embodiments, the detection object includes a flow corresponding to a flow affinity table, and the system measures the detection object according to the computing network operation and maintenance management component 150 and the computing power routing and forwarding component 140, including:

[0092] The computing power routing forwarding component 140 receives the computing power routing table determined by the computing power routing calculation component 130 according to the computing power measurement index information and the network measurement index information; the computing power routing forwarding component 140 deployed at the inlet gateway receives the computing power service access request of the terminal user, determines the flow affinity table according to the computing power service access request and the computing power routing table, and sends the enabling information to the computing network operation and maintenance management component 150; the computing network operation and maintenance management component 150 constructs the operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing forwarding component 140 deployed at the inlet gateway; the computing power routing forwarding component 140 deployed at the inlet gateway encapsulates and forwards the operation and maintenance management message to the computing network operation and maintenance management component 150 deployed at the egress gateway. The computing power routing forwarding component 140 deployed at the egress gateway decapsulates the operation and maintenance management message, sends the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, encapsulates the operation and maintenance management message response and forwards it to the computing power routing forwarding component 140 deployed at the ingress gateway; the computing power routing forwarding component 140 deployed at the ingress gateway decapsulates the operation and maintenance management message response, and sends the operation and maintenance management message response to the computing network operation and maintenance management component 150; the computing network operation and maintenance management component 140 detects the flow affinity table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing computing component 130.

[0093] In an embodiment of the present application, the computing power routing calculation component 130 can determine the computing power routing table through computing power measurement index information and network measurement index information, and send the computing power routing table to the computing power routing forwarding component 140. The terminal user can send a computing power service access request, and the computing power routing forwarding component 140 deployed on the inlet gateway can receive the above computing power service access request. The computing power routing forwarding component 140 can create a flow affinity table according to the computing power routing table and the computing power service access request. For example, the computing power routing forwarding component 140 can use the computing power identifier and the source Internet Protocol (IP) address carried by the computing power service request as key values, and search the computing power routing table for the corresponding path according to the key value. If the path is found, a flow affinity table is created. If the path is not found, the computing power service request is discarded. In the created flow affinity table, the computing power identifier and the source IP are key values, and the corresponding segment routing (SR) policy, the egress gateway IP address, and the computing instance IP address in the computing power routing table can be used as attributes of the flow affinity table.

[0094] Specifically, the computing power routing forwarding component 140 deployed at the inlet gateway can also send enabling information to the computing network operation and maintenance management component 150, and the enabling information can instruct the computing network operation and maintenance management component 150 to detect the flow corresponding to the created flow affinity table. After receiving the enabling information, the computing network operation and maintenance management component 150 can encapsulate the operation and maintenance management message and send the encapsulated computing network operation and maintenance management message to the computing power routing forwarding component 140 deployed at the inlet gateway, and the computing power routing forwarding component 140 deployed at the inlet gateway encapsulates the computing network operation and maintenance management message and forwards the encapsulated operation and maintenance management message to the computing power routing forwarding component 140 deployed at the egress gateway. Component 140 can decapsulate the operation and maintenance management message and send the operation and maintenance management message to the computing instance to obtain the operation and maintenance management message response of the computing instance in response to the maintenance operation management message. The computing power routing and forwarding component 140 deployed at the egress gateway can encapsulate the operation and maintenance management message response and transmit it to the computing power routing and forwarding component 140 deployed at the ingress gateway. The computing power routing and forwarding component 140 deployed at the ingress gateway decapsulates the operation and maintenance management message response and transmits the operation and maintenance management message response to the computing network operation and maintenance management component 150. The computing network operation and maintenance management component 150 can detect the corresponding flow in the flow affinity table based on the operation and maintenance management message response and provide feedback on the generated detection results.

[0095] In some application embodiments, the system performs network operation, maintenance, and management measurement feedback based on the computing power sensing component 120 and the computing power routing calculation component 130. The network operation, maintenance, and management measurement feedback includes at least one of the following:

[0096] The computing power routing calculation component 130 receives the detection result, recalculates and corrects the detection object corresponding to the detection result;

[0097] The computing power sensing component 120 receives the detection result and notifies the detection result to the protocol diffusion.

[0098] In an embodiment of the present application, the computing power routing calculation component 130 can receive the detection results fed back by the computing network operation and maintenance management component 150. The computing power routing calculation component 130 can recalculate the corresponding detection object based on the detection results to modify the service level agreement target of the detection object. Alternatively, the computing power perception component 120 can receive the detection results fed back by the computing network operation and maintenance management component 150. The computing power perception component 120 can notify the protocol of the detection results, disseminate the detection results through the notification protocol, update the computing power status database, and trigger the upstream computing network routing calculation component to recalculate the computing power routing.

[0099] Based on the above application embodiment, the detection object includes at least one of the computing power routing table and the flow affinity table.

[0100] In an embodiment of the present application, the detection result received by the computing power routing calculation component 130 may be the result of detecting the flow corresponding to the computing power routing table or the flow affinity table, and the computing power routing calculation component 130 may recalculate the flow corresponding to the computing power routing table or the flow affinity table.

[0101] Based on the above application embodiment, the computing network operation and maintenance management component 150 includes at least one of the following: a computing network operation and maintenance management configuration module and a computing network operation and maintenance management execution module.

[0102] In an embodiment of the present application, the computing network operation and maintenance management component 150 can be composed of at least one of a computing network operation and maintenance management module and a computing network operation and maintenance management execution module, wherein the computing network operation and maintenance management module can be responsible for managing the configuration enablement objects issued by the OAM application component, and the OAM application component includes but is not limited to a network perception component, a computing power perception component, a computing power routing calculation component, etc. The OAM application component can also include a subscription initiator or an OAM detection result user that is enabled by the configuration of the OAM detection object, for example, a user terminal that initiates a computing power service access request. The computing network operation and maintenance management configuration module can initiate OAM detection instances for detection objects with different configurations, and receive the detection results of the OAM application component, and publish the detection results according to the detection object enabling subscription demand side to trigger network convergence or operation indicator visualization. The computing network operation and maintenance management execution module can execute OAM instructions on the detection object, encapsulate and send detection messages to the computing power routing forwarding component, receive OAM response messages sent by the computing power routing forwarding component, and trigger the detection result reporting feedback based on the detection results and thresholds.

[0103] In an exemplary embodiment, referring to FIG3 , the computing network operation and maintenance management system can cooperate with the cloud resource pool to implement OAM detection of detection objects of different granularities, thereby improving the consistency of computing network services. The overall architecture of the computing network operation and maintenance management system and the cloud resource pool may include the following:

[0104] 1. Cloud resource pool: The cloud resource pool includes physical computing hosts and is typically virtualized to achieve resource pooling. The cloud resource pool is used to deploy computing instances. Typically, a cloud-side computing power collection component is deployed in the cloud resource pool to collect metrics for each computing instance. Key metrics include processing latency, CPU usage, memory usage, and bandwidth usage. This is communicated to the computing power gateway device via a common protocol to achieve network computing power awareness.

[0105] 2. Network awareness components: These are typically deployed on the control plane and primarily include IGP / BGP and Telemetry. They enable awareness and information dissemination of network node and link quality (latency, bandwidth, bandwidth usage, traffic information, etc.).

[0106] 3. Computing power perception component: In the embodiment of the present application, the computing power perception component is deployed on the control plane and connected to the cloud-side computing power collection component to realize the perception of raw computing power information. At the same time, the computing power perception component can also be deployed on upstream network nodes or network controllers to receive and disseminate computing power information. Since the computing power status has a strong time-varying nature, the advertising interval of the computing power perception component is usually kept at the same level as the routing protocol to reduce the pressure on the control plane;

[0107] 4. Computing power routing calculation component: The computing power routing calculation component is deployed on the control plane, connecting to the network and computing power perception components respectively, obtaining and summarizing network and computing power related metrics information. The computing power routing calculation component is used to dynamically calculate computing power routes and send computing power routes to the computing power routing forwarding component;

[0108] 5. Computing power routing and forwarding component: To support efficient service forwarding, the computing power routing and forwarding component is usually deployed on the data plane to receive computing power routing information issued by the control plane. The computing power routing and forwarding component can provide integrated computing and network forwarding services by identifying service messages carrying ServiceID.

[0109] 6. Computing network OAM component: The overall function of the computing network OAM component is to complete the detection of indicators including but not limited to connectivity failures, packet loss and delay. Depending on the different OAM detection objects, the computing network OAM component in the present application example can simultaneously receive the detection objects of the computing power perception component and the computing power routing calculation component, complete OAM detection on the detection objects and report the results.

[0110] In Figure 3, the functions of the interfaces between different components can be shown in Table 1:

[0111] Table 1 Interface function table

[0112] In the system operation process shown in Figure 3, OAM management can be achieved through the following functions:

[0113] 1. Computing network status collection: This involves the cloud-side computing power collection component, computing power perception component, and network perception component. The computing network status collection process primarily involves: 1) The cloud-side computing power collection component notifies the collected computing power information to the computing power perception component for dissemination across the network domain, forming a computing power status database; 2) The network domain, through the network perception component, typically using IGP / BGP-LS / Telemetry, collects network status information to form a network status database.

[0114] 2. Joint computing of computing power and network: involves computing power routing computing components and computing power routing forwarding components. The computing network joint computing process typically includes the following: 1) The computing power routing computing component performs joint computing based on the computing power status and network status, selects a specific service instance and the corresponding network path, and sends the result (computing power routing table) to the computing power routing forwarding component; 2) The computing power routing forwarding component searches the computing power routing table for the service identifier carried in the terminal request and forms a flow affinity table. This flow affinity table is used to ensure that the user session always accesses the same computing instance during the connection lifecycle to avoid session interruption; 3) The flow affinity table is usually also sent to the computing power routing computing component, which manages and maintains the connected user session; 4) Based on the computing power routing OAM and service flow OAM configuration enabled on the management plane, the detection object is sent to the computing network OAM component, and the computing network OAM component is activated for the computing network OAM detection capability of the detection object. The computing power routing computing component receives the detection result feedback and analyzes the service level agreement (SLA) target gap of the computing power routing and service flow services based on the detection results, adjusts the detection object, and continuously corrects the network path and / or computing instance to ensure the target is achieved.

[0115] 3. Computing network OAM measurement: involves computing network OAM components, computing power routing and forwarding components, and computing instances. In order to implement OAM measurement, OAM measurement tools are usually deployed on computing instances. For example, OAM measurement tools include ping, twamp, owamp, flow detection, etc. or their combination. The computing network OAM measurement process usually includes: 1) The computing network OAM component initiates an OAM message according to the configured detection object and detection protocol; 2) At the ingress gateway, the computing power routing and forwarding component queries the forwarding table to encapsulate the segment routing header (SRH) tunnel, performs OAM detection on the computing power routing, and usually also encapsulates the computing instance IP. At the egress gateway, the tunnel is unpacked and the detection message is restored; 3) The detection message is delivered to the computing instance measurement tool, and the measurement result is returned to the computing network OAM component.

[0116] 4. Computing network OAM measurement feedback: The computing network OAM component completes feedback transmission, and the computing power perception component and computing network computing component complete feedback result reception and processing. The process generally includes: 1) The computing network OAM component compares the OAM measurement results, identifies the corresponding detection object and configured threshold based on the detection results, and compares the configured threshold with the OAM measurement result to determine whether to provide feedback; 2) If the computing power perception component deployed on the egress gateway receives the detection result of the computing instance detection, it performs computing power diffusion on the detection result to trigger the computing power routing calculation component deployed on the ingress gateway or centralized computing network controller to perform calculation immediately; 3) If the computing power routing calculation component receives the computing power routing table detection result, if the detection result does not meet the computing network SLA target expectation, it immediately calculates and refreshes the result to generate a new computing power routing table and sends it to the computing power routing forwarding component. If the service flow detection result is received, based on the gap between the service SLA target and the actual measurement, it triggers the network to perform compensation path calculation according to the gap target, generates a new path and sends it to the computing power routing forwarding component to update the service flow forwarding path.

[0117] In the embodiment of the present application, the slow lane control surface is slow collected / measured + calculated to form a service forwarding plan through the above 1 and 2, and the basic service action control is completed through the issuance of rules. The fast lane is fast collected / measured and the correction / delimitation of the calculation is achieved through the above 3 and 4, and the service is optimized through control. The entire computing network system is a negative feedback closed-loop system, which can better provide a computing network service with a consistent experience.

[0118] Referring to Figure 4 , this application adds computing network OAM-related components and defines related interfaces with the computing network gateway as the center to complete the computing network OAM functions and provide services. It includes the following main parts:

[0119] 1. Configuration management component: Perform global function configuration, implement the overall function OAM northbound interface, and support configuration of computing network OAM on or off.

[0120] 2. OAM application components: mainly include network perception components and computing components in the overall computing network OAM architecture, and can also be other application components. They are the initiators of configuration, enablement, subscription and (or) users of OAM detection results of computing network OAM detection objects.

[0121] 3. Computing Network OAM Management Module: This module manages the configuration enablement objects issued by the OAM application component, initiates OAM detection instances for different configuration objects and demand levels, receives the results of the OAM execution components, and publishes the detection results to the target subscribers based on the object enablement to trigger network convergence or visualize operational indicators.

[0122] 4. Computing network OAM execution module: executes OAM instructions on the detection object, encapsulates and sends detection messages to the computing power routing forwarding component, receives OAM response messages sent by the computing power routing forwarding component. Its interface refers to interface X8 in the overall computing network OAM architecture, and organizes the detection results and reports them to the management component based on the threshold trigger. The accelerated forwarding plane hardware can include field programmable gate arrays (FPGAs), dedicated network processors (NPs), etc., which can realize the separation of fast lane execution and slow lane management.

[0123] 5. Computing power routing forwarding component: Based on the detection message sent by the OAM execution module, it searches at least one of the flow affinity table, computing power routing, and IP routing table in the outbound direction according to the message and configuration instructions at the forwarding layer to further encapsulate and forward the OAM message. At the same time, it receives and processes the OAM response and extracts it and sends it to the OAM execution module.

[0124] In Figure 4, the key interfaces in the computing network OAM architecture are described as follows:

[0125] 1. Function configuration: This is mainly configured by the northbound interface to enable or disable the computing network OAM function. Specifically, the enable / disable function can be targeted at the entire network, computing instances, computing power routing tables, and business flows.

[0126] 2. Object enablement subscription: For three object types, namely computing instances, computing power routing tables, and business flows, OAM is enabled for specific entries. Specifically, the enabling message for computing instances includes but is not limited to: source IP, computing instance IP, detection type, detection period, and reporting threshold; the enabling message for computing power routing table instances or business flow instances includes but is not limited to: source IP, ServiceID, detection type, detection period, and reporting threshold.

[0127] 3. Release of test results: Based on the test results, the test results are sent to the application module for the object enable subscription information. Refer to the object enable subscription interface. The reporting interface usually includes but is not limited to source IP, destination IP, connectivity, packet loss, latency, bandwidth, etc.

[0128] 4. Internal interface: In the downstream direction, the OAM management module classifies, aggregates, and deduplicates the object enablement subscription information and sends it to the execution component for execution. In the upstream direction, the OAM execution module sends the detection results based on the threshold breakthrough according to the detection object. Therefore, the upstream and downstream interfaces are basically the same as the above two steps.

[0129] 5. Internal transmission and reception of detection messages: Depending on the selected OAM protocol, the internal transmission and reception messages vary. Relevant detection protocols may include but are not limited to ping, twamp, owamp, and stamp. New detection protocols can also be expanded to cover detection indicators such as FM, PM, and swap management (XM).

[0130] 6. Detect message transmission and reception at the network level: Depending on the selected OAM protocol, external message transmission and reception, and related detection protocols may include but are not limited to ping, twamp, owamp, and stamp. Unlike the previous step, these protocol messages may be encapsulated in a tunnel depending on the detection level and restored to IP OAM messages after being decapsulated by the egress gateway to reach the computing instance.

[0131] FIG5 is a flowchart of a computing network operation and maintenance management method provided in an embodiment of the present application. The embodiment of the present application is applicable to situations where a computing network operation and maintenance management system performs computing network operation and maintenance management, including OAM detection of computing instances. The computing network operation and maintenance management method provided in the embodiment of the present application can be implemented by software and / or hardware. Referring to FIG5 , the method provided in the embodiment of the present application specifically includes the following steps:

[0132] Step 210: The computing power perception component obtains the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and sends enabling information to the computing network operation and maintenance management component based on the computing power information to start the computing network operation and maintenance management component.

[0133] In an embodiment of the present application, the computing power perception component can be connected to the cloud-side computing power collection component deployed in the cloud resource pool, and can obtain the computing power information collected by the cloud-side computing power collection component. The computing power perception component can generate enabling information when obtaining the computing power information. The enabling information can trigger the computing network management and maintenance operation component to detect the computing instance deployed in the cloud resource pool based on the computing power information. It can be understood that the enabling information can indicate that the object of OAM detection is the computing instance, and the computing power perception component can send the enabling information to the computing network operation and maintenance management component.

[0134] Step 220: The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component.

[0135] Specifically, upon receiving the enabling information, the computing network operation and maintenance management component can assemble an operation and maintenance management message. This operation and maintenance management message can be used to detect the computing instance. This operation and maintenance management message can be a computing instance-level message. The computing network operation and maintenance management component can assemble the operation and maintenance management message according to its configured protocol format. The computing network operation and maintenance management component can transmit the assembled operation and maintenance management message to the computing power routing and forwarding component.

[0136] Step 230: The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, so that the computing instance generates an operation and maintenance management message response according to the deployed operation and maintenance management service and the operation and maintenance management message protocol.

[0137] In an embodiment of the present application, the computing power routing forwarding component can forward the operation and maintenance management message for the computing instance, thereby sending the operation and maintenance management message to the computing instance. The computing instance can respond to the operation and maintenance management message and generate an operation and maintenance management message response.

[0138] Step 240: The computing power routing forwarding component receives the operation and maintenance management message response of the computing instance, and sends the operation and maintenance management message response to the computing network operation and maintenance management component.

[0139] In an embodiment of the present application, the computing instance can generate an operation and maintenance management message response and send it to the computing power routing forwarding component. The computing power routing forwarding component can receive the operation and maintenance management message and forward the operation and maintenance management message to the computing network operation and maintenance management component.

[0140] Step 250: The computing network operation and maintenance management component generates an operation and maintenance management detection result according to the operation and maintenance management message response, and determines an operation and maintenance management detection notification message according to the detection object information.

[0141] In an embodiment of the present application, the computing network operation and maintenance management component can detect the computing power performance indicators of the computing instance based on the operation and maintenance management message response. The computing power performance indicators may include but are not limited to the computing instance IP, connectivity, packet loss, and latency. The detected computing power performance indicators can be used as the operation and maintenance management detection results. The computing network operation and maintenance management component can determine the detection object information, that is, the operation and maintenance management result is the indication information for detecting the computing instance. The operation and maintenance management detection result can be processed into an operation and maintenance management detection notification message according to the detection object information, so that the operation and maintenance management detection notification message can notify the operation and maintenance management detection result within the system.

[0142] Step 260: The computing power perception component receives the operation, maintenance, and management detection notification message and spreads the computing power information corresponding to the operation, maintenance, and management detection notification message.

[0143] Specifically, when the computing power perception component receives an operation and maintenance management detection notification message that notifies the operation and maintenance management detection result, the computing power information corresponding to the operation and maintenance management detection notification message can be diffused. It can be understood that the computing power information can be the computing power information of the computing instance corresponding to the operation and maintenance management detection result notified by the diffused operation and maintenance management detection notification message.

[0144] Step 270: The upstream node receives and processes the computing power information, recalculates the computing power route, and sends the computing power route to the forwarding component.

[0145] In an embodiment of the present application, the upstream node can receive the computing power information, recalculate the computing power route of the computing power information, and send the recalculated computing power route to the computing power route forwarding component.

[0146] In some application embodiments, it also includes: the computing network operation and maintenance management component determines the detection object based on the operation and maintenance management detection result, and determines the feedback notification based on the configuration threshold of the detection object and the operation and maintenance management detection result.

[0147] The configured threshold value may be a pre-configured minimum indicator value for notifying the detection result, and the configured threshold value may be an indicator value or an indicator value set.

[0148] In an embodiment of the present application, the corresponding detection object can be determined through the operation and maintenance management detection result, the configuration threshold value associated with the detection object can be found, and the operation and maintenance management detection result can be compared with the corresponding preset threshold value to determine whether to feedback the operation and maintenance management detection result. If so, an operation and maintenance management detection notification message is generated according to the operation and maintenance management detection result. The operation and maintenance management detection notification message can be used to notify the operation and maintenance management detection result. If not, the operation and maintenance management detection result may not be notified.

[0149] Figure 6 is an example diagram of a computing network operation and maintenance management method provided by an embodiment of the present application. It uses computing instance-level OAM and application processes as examples to introduce the implementation of computing power routing. The computing instance-level information is generally perceived by the egress gateway and spread upstream. The distributed head node or centralized control plane performs computing power routing calculations. The embodiment of the present application initiates instance-level OAM for the computing instance perceived by the egress gateway to detect the connectivity, packet loss, comprehensive delay and other information of the computing instance. It can trigger the control plane to converge quickly and provide more accurate services. As shown in Figure 6, the main process is described as follows:

[0150] S1: Cloud resource deployment uses cloud-side computing power resource collection components to collect computing power information. To reduce control plane pressure, protocols such as the Border Gateway Protocol (BGP) are usually extended to use a slow cycle to report computing power information.

[0151] S2: The computing power perception component sends an enable message to the computing network OAM component to start instance OAM based on the computing power information reported by S1 above. The computing power information includes the computing instance IP information, the connected interface IP, the detection period configured for the northbound interface, etc.

[0152] S3: The computing network OAM component assembles an OAM message based on the enabling information. The source IP included in the OAM message is the interface IP and the destination IP is the corresponding computing instance IP. The OAM message is sent to the computing power routing forwarding component. The OAM message can be ping, twamp, owamp, stamp, etc. or a combination of multiple protocols (including extensions).

[0153] S4: The egress gateway forwards the computing instance-level OAM message through the computing power routing forwarding component, passing through the data communication network (DCN) and the physical server network card to the cloud-side computing instance.

[0154] S5: The cloud-side computing instance generates an OAM message response based on the deployed OAM service and OAM message protocol. The OAM message response carries information such as packet loss and latency. The destination IP address of the OAM message response is the source IP address of the OAM message, and the source IP address of the OAM message response is the computing instance IP address.

[0155] S6: The computing routing and forwarding component of the egress gateway receives the OAM message response, identifies the OAM message as destined for the egress gateway interface IP address, and sends the OAM message response to the computing network OAM component. This corresponds to Figure 6. At this point, the computing network OAM component has successfully detected failures, packet loss, and latency between the egress gateway and the computing instance.

[0156] S7: The computing network OAM component compares the detection results based on the enabled detection object information, such as the computing instance IP, connectivity, packet loss, latency, and other indicators, to decide whether to notify the computing power perception component of the detection results. The decision criteria are usually: whether there is a fault, whether the packet loss threshold is exceeded, whether the latency threshold is exceeded, and other conditions or combinations. The conditions depend on the configuration of the detection object and related parameters.

[0157] S8: The computing power perception component receives the OAM detection notification message and immediately triggers and spreads the corresponding computing power information.

[0158] S9-S10: The upstream node receives and processes relevant computing power information, triggers the recalculation of computing power routing and sends down updates, making up for the lack of computing power measurement range and perception real-time performance, and avoiding computing power routing black holes.

[0159] FIG7 is a flowchart of another computing network operation and maintenance management method provided by an embodiment of the present application. The embodiment of the present application is applicable to situations where a computing network operation and maintenance management system performs computing network operation and maintenance management, including OAM detection at the computing power routing level. The computing network operation and maintenance management method provided by the embodiment of the present application can be implemented by software and / or hardware. Referring to FIG7 , the method provided by the embodiment of the present application specifically includes the following steps:

[0160] Step 310: The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component.

[0161] In an embodiment of the present application, the network perception component can send the measurement process information it perceives to the computing power routing calculation component, and the computing power perception component can send the measurement process information it perceives to the computing power routing calculation component, and the computing power routing calculation component can receive the above-mentioned measurement process information.

[0162] Step 320: The computing power routing calculation component generates a computing power routing table based on the measurement process information, and sends the computing power routing table to the computing power routing forwarding component.

[0163] In an embodiment of the present application, the computing power routing calculation component can process the measurement process information to generate a computing power routing table, and send the generated computing power routing table to the computing power routing forwarding component. It can be understood that the calculation process of the computing power routing table can be triggered by an event or a periodic timing trigger.

[0164] Step 330: The computing power routing calculation component sends enabling information to the computing network operation and maintenance management component according to the computing power routing table.

[0165] In an embodiment of the present application, when generating a new computing power routing table, the computing power routing calculation component can send enabling information to the computing network operation and maintenance management component. The enabling information can trigger the computing network operation and maintenance management component to perform OAM detection on the newly generated computing power routing table.

[0166] Step 340: The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component.

[0167] Specifically, when the computing network operation and maintenance management component receives the enabling information, it can assemble the corresponding operation and maintenance management message according to the detection object indicated by the enabling information, that is, assemble the operation and maintenance management message for detecting the computing power routing table. The computing network operation and maintenance management component can send the assembled operation and maintenance management message to the computing power routing forwarding component for forwarding.

[0168] Step 350: The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component.

[0169] In an embodiment of the present application, the computing power routing forwarding component can forward the operation and maintenance management message so that the operation and maintenance management message reaches the computing instance. The computing instance can respond to the operation and maintenance management message and generate an operation and maintenance management message reply. The computing instance can send the generated operation and maintenance management message reply to the computing power routing forwarding component. The computing power routing forwarding component can receive the operation and maintenance management message and forward the operation and maintenance management message to the computing network operation and maintenance management component.

[0170] Step 360: The computing network operation and maintenance management component generates a detection result of the computing power routing table based on the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component.

[0171] In an embodiment of the present application, the computing network operation and maintenance management component can detect the computing power performance indicators of the computing power routing table based on the operation and maintenance management message response. The computing power performance indicators may include but are not limited to the computing instance IP, connectivity, packet loss, and latency. The detected computing power performance indicators can be used as detection results. The computing network operation and maintenance management component can also feed back the generated detection results to the computing power routing calculation component. It is understood that before the detection result is fed back, it can also be determined whether to feedback based on the configured threshold threshold of the corresponding detection object.

[0172] Step 370: The computing power routing calculation component determines the service level agreement target gap based on the detection results, recalculates the computing power routing table based on the service level agreement target gap, and sends the computing power routing table to the computing power routing forwarding component.

[0173] Specifically, the computing power routing calculation component can compare the computing power performance index of the detection result with the computing power performance index of the service level agreement target, determine the service level agreement target gap between the detection result and the service level agreement target, and recalculate the computing power routing table according to the service level agreement target gap to reduce the service level agreement target gap. The computing power routing calculation component can re-send the regenerated computing power routing table to the computing power routing forwarding component. It can be understood that in some application embodiments, the computing power routing calculation component can recalculate certain computing power routes of the computing power routing table.

[0174] Based on the above application embodiment, the computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information and sends the operation and maintenance management message to the computing power routing forwarding component, including:

[0175] The computing network operation and maintenance management component assembles the operation and maintenance management message according to the enabling information, wherein the message format of the operation and maintenance management message is determined according to the operation and maintenance management protocol selected by the computing network operation and maintenance management component; the computing network operation and maintenance management component sends the operation and maintenance management message to the computing power routing forwarding component deployed at the ingress gateway.

[0176] In an embodiment of the present application, the computing network operation and maintenance management component can determine the operation and maintenance management protocol selected by the computing network operation and maintenance management component according to the detection object corresponding to the enabling information, can determine the message format of the operation and maintenance management message according to the detection object and the operation and maintenance management protocol selected by the computing network operation and maintenance management component, and can assemble the operation and maintenance management message according to the message format; the computing network operation and maintenance management component can forward the assembled operation and maintenance management message to the computing power routing forwarding component deployed at the ingress gateway.

[0177] Based on the above application embodiment, the computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component, including:

[0178] The computing power routing forwarding component deployed at the ingress gateway encapsulates the received operation and maintenance management message, and forwards the operation and maintenance management message to the computing power routing forwarding component deployed at the egress gateway; the computing power routing forwarding component deployed at the egress gateway decapsulates the operation and maintenance management message, sends the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, encapsulates the operation and maintenance management message response and forwards it to the computing power routing forwarding component deployed at the ingress gateway; the computing power routing forwarding component deployed at the ingress gateway decapsulates the operation and maintenance management message response, and sends the operation and maintenance management message response to the computing network operation and maintenance management component.

[0179] In an embodiment of the present application, the computing power routing forwarding component deployed at the ingress gateway encapsulates the computing network operation and maintenance management message, and forwards the encapsulated operation and maintenance management message to the computing power routing forwarding component deployed at the egress gateway. The computing power routing forwarding component deployed at the egress gateway can decapsulate the operation and maintenance management message, and send the operation and maintenance management message to the computing instance to obtain the operation and maintenance management message response of the computing instance in response to the maintenance operation management message. The computing power routing forwarding component deployed at the egress gateway can encapsulate the operation and maintenance management message response and transmit it to the computing power routing forwarding component deployed at the ingress gateway. The computing power routing forwarding component deployed at the ingress gateway decapsulates the operation and maintenance management message response and transmits the operation and maintenance management message response to the computing network operation and maintenance management component. It can be understood that the operation and maintenance management message and the operation and maintenance management message response can be encapsulated and decapsulated by the computing power routing forwarding components deployed at the egress gateway and the ingress gateway, so as to realize full-level transmission of messages and unified management of OAM detection messages.

[0180] Figure 8 is an example diagram of a computing network operation and maintenance management method provided by an embodiment of the present application. Taking the computing power routing level OAM and application process as an example, the computing power routing is generally realized by sensing the computing instance level information through the egress gateway and spreading it upstream. The ingress gateway or centralized controller controls the computing power routing calculation, and detects the service path quality of the traffic engineering (TE) path (such as SR-Policy) calculated by the network through OAM technologies such as SR-PM and realizes a closed loop of network service quality. Computing power routing provides a comprehensive selection of computing instances and paths, and performs OAM detection on computing power routing table entries to detect connectivity, packet loss, comprehensive delay and other information of the service routing table, evaluate the achievement of computing power routing network SLA goals, and can also trigger the control plane to converge quickly based on the gap, providing more accurate and timely service capabilities. As shown in Figure 8, the main processes are described as follows:

[0181] S1-S2: The network and computing power perception components send metric information to the computing power routing calculation component respectively.

[0182] S3: The computing power routing calculation component receives the network and computing power metrics, triggers computing power routing calculation based on events or periodic timing, and sends the computing power routing table to the computing power routing forwarding component to perform subsequent computing power service business forwarding.

[0183] S4: The computing power routing calculation component sends the computing power routing entry OAM enablement information to the computing network OAM component based on the computing power routing entry. The key elements of the OAM enablement information include: source IP, computing power ID, and detection period. In particular, the source IP is the IP address of any interface on the ingress gateway of the VRF where the computing power routing table is located.

[0184] S5: The computing network OAM component assembles an OAM message based on the configured enable information and sends it to the computing power routing and forwarding component. The encapsulation format of the OAM message is determined by the selected OAM protocol. Its destination IP is the computing power identifier and the source IP is any interface IP address mentioned in S4.

[0185] S6: The ingress gateway computing power routing forwarding component searches the computing power routing table based on the destination address of the OAM message to obtain information such as the SR-Policy, egress gateway, and computing instance, encapsulates the tunnel header and SRH information, and forwards the OAM message.

[0186] S7: The egress gateway computing power routing forwarding component decapsulates the tunnel header and SRH, and further forwards the OAM message to the computing instance based on the computing instance IP.

[0187] S8: The computing instance responds to the OAM message with information such as packet loss and latency based on the deployed OAM service and OAM message protocol. The destination IP address of the response message is the source IP address of the OAM message, and the source IP address of the response message is the computing instance IP address.

[0188] S9: The computing instance OAM response message received by the egress gateway computing power routing forwarding component is encapsulated and forwarded in the downlink tunnel.

[0189] S10: The ingress gateway computing power routing and forwarding component receives the computing network OAM response message, removes the tunnel and SRH encapsulation, matches the local IP address based on the OAM response message's destination IP address, and sends the message to the computing network OAM component. This corresponds to Figure 8 . At this point, computing network OAM is used to evaluate the computing network's SLA targets for the computing power routing table. These SLA targets can include computing network packet loss, computing network latency, and other computing network metrics.

[0190] S11: If the computing network OAM component compares the detection object mapped by the OAM message and the fault detection times out or the packet loss and delay statistics exceed the threshold, the computing power routing calculation component is notified to recalculate the detection object.

[0191] S12: The computing power routing calculation component decides whether to recalculate the computing power routing table for a specific ServiceID based on the measurement results and the target SLA gap, and sends it to the computing power routing forwarding component for computing power services. At the same time, the computing power routing OAM process is also executed. At this point, the S4-S12 process is repeated in a closed loop to keep the computing power routing calculation result relatively optimal.

[0192] FIG9 is a flowchart of another computing network operation and maintenance management method provided in an embodiment of the present application. The embodiment of the present application is applicable to a computing network operation and maintenance management system for performing computing network operation and maintenance management, including OAM detection at the computing power service session level. The computing network operation and maintenance management method provided in the embodiment of the present application can be implemented by software and / or hardware. Referring to FIG9 , the method provided in the embodiment of the present application specifically includes the following steps:

[0193] Step 410: The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component.

[0194] In an embodiment of the present application, the network perception component can send the measurement process information it perceives to the computing power routing calculation component, and the computing power perception component can send the measurement process information it perceives to the computing power routing calculation, and the computing power routing calculation component can receive the above-mentioned measurement process information.

[0195] Step 420: The computing power routing calculation component generates a computing power routing table based on the measurement process information, and sends the computing power routing table to the computing power routing forwarding component.

[0196] In an embodiment of the present application, the computing power routing calculation component can process the measurement process information to generate a computing power routing table, and send the generated computing power routing table to the computing power routing forwarding component. It can be understood that the calculation process of the computing power routing table can be triggered by an event or a periodic timing trigger.

[0197] Step 430: The computing power routing forwarding component determines the flow affinity table based on the received computing power routing table and the computing power service access request of the terminal user, and sends enabling information to the computing network operation and maintenance management component.

[0198] In an embodiment of the present application, a terminal user can send a computing power service access request according to their own needs. The computing power routing forwarding component can determine the flow affinity table according to the computing power routing table it receives and the computing power service access request of the terminal user. For example, it extracts the computing power identifier and source IP address carried in the computing power service access request, and searches for a corresponding flow affinity table according to the computing power identifier and source IP. If it exists, the flow affinity table is used as the determined flow affinity table. If it does not exist, the computing power routing table is searched according to the computing power identifier to confirm whether there is a computing power route corresponding to the computing power identifier. If it exists, a new flow affinity table is created as the determined flow affinity table. Otherwise, the computing power service access request is discarded. The computing power routing forwarding component can generate enabling information according to the determined flow affinity table. The enabling information can instruct the computing network operation and maintenance management component to perform OAM detection on the flow corresponding to the determined flow affinity table. The computing power routing forwarding component sends the enabling information to the computing network operation and maintenance management component.

[0199] Step 440: The computing network operation and maintenance management component constructs an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component.

[0200] Specifically, when the computing network operation and maintenance management component receives the enabling information, it can assemble the corresponding operation and maintenance management message according to the detection object indicated by the enabling information, that is, assemble the operation and maintenance management message used to detect the flow corresponding to the flow affinity table. The computing network operation and maintenance management component can send the assembled operation and maintenance management message to the computing power routing forwarding component for forwarding.

[0201] Step 450: The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component.

[0202] In an embodiment of the present application, the computing power routing forwarding component can forward the operation and maintenance management message so that the operation and maintenance management message reaches the computing instance. The computing instance can respond to the operation and maintenance management message and generate an operation and maintenance management message reply. The operation and maintenance management message reply generated by the computing instance can be sent to the computing power routing forwarding component. The computing power routing forwarding component can receive the operation and maintenance management message and forward the operation and maintenance management message to the computing network operation and maintenance management component.

[0203] Step 460: The computing network operation and maintenance management component determines the detection result based on the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component.

[0204] In an embodiment of the present application, the computing network operation and maintenance management component can detect the computing power performance indicators of the computing power routing table based on the operation and maintenance management message response. The computing power performance indicators may include but are not limited to computing instance IP, connectivity, packet loss, latency, and other computing network indicators. The detected computing power performance indicators can be used as detection results. The computing network operation and maintenance management component can also feed back the generated detection results to the computing power routing calculation component. It is understood that before the detection result is fed back, the configuration threshold threshold of the corresponding detection object can also be used to determine whether to feedback.

[0205] Step 470: The computing power routing calculation component determines the service level agreement target gap corresponding to the detection result, calculates the path compensation based on the service level agreement target gap, and sends the path compensation to the computing power routing forwarding component to refresh the flow affinity table.

[0206] In an embodiment of the present application, the computing power routing calculation component can compare the computing power performance index of the detection result with the computing power performance index of the service level agreement target to determine the service level agreement target gap between the detection result and the service level agreement target. The computing power routing calculation component can determine a path compensation based on the service level agreement target gap and transmit the path compensation to the computing power routing forwarding component, so that the computing power routing forwarding component updates the stored flow affinity table according to the above path compensation.

[0207] Figure 10 is an example diagram of a computing network operation and maintenance management method provided by an embodiment of the present application. Taking computing service session-level OAM and applications as an example, since computing routing is highly time-varying, a flow affinity table is formed on the data plane of a specific computing routing entry when the user's first packet accesses it. The network SR-Policy path in the corresponding routing table is bound to the flow affinity table, and subsequent packets are forwarded based on the flow affinity table. The flow affinity table usually uses a five-tuple as a key value to identify flow information to perform subsequent tunnel and SRH encapsulation processing and forwarding. As the access load of the computing instance terminal increases, its processing delay quality gradually deteriorates. If the SR-Policy path bound to the flow affinity table is not detected in time and dynamically adjusted to compensate, the terminal user access QOE will not be guaranteed. Based on the terminal source IP and computing power identifier as the key value of the session-level flow affinity table, OAM detection is performed on the flow affinity session to check end-to-end connectivity, packet loss, comprehensive latency and other information, and evaluate the achievement of QOE goals. At the same time, based on the gap in the computing network SLA target, the control plane is triggered to recalculate the new SR-Policy and send it to the data plane and bind it to the flow affinity table to maintain service continuity and experience consistency. As shown in Figure 10, the main process is described as follows:

[0208] S1-S2: The network and computing power perception components respectively send computing power metric information to the computing power routing calculation component.

[0209] S3: The computing power routing calculation component receives network and computing power metric information, triggers computing power routing calculation according to events or periodic timing, and sends the computing power routing table to the computing power routing forwarding component to perform subsequent computing power service business forwarding.

[0210] S4: The end user obtains the computing power identifier through the Domain Name System (DNS) or manual configuration, and initiates access to the computing power service. Usually, the computing power identifier is carried in the destination address as an anycast address or in the extension header (EH).

[0211] S5: The computing power routing forwarding component of the ingress gateway receives the end-user request. Its search process: first, based on the computing power ID and source IP address, it searches the flow affinity table. If the flow affinity table does not exist, it searches the computing power routing table instead. If it is found, it creates the flow affinity table; otherwise, it discards the message. The newly created flow affinity table uses the computing power ID and source IP address as key values, and the corresponding SR-Policy, egress gateway IP address, and compute instance IP in the computing power routing table as corresponding attributes. This allows packets matching the flow affinity table to be tunneled and SRH encapsulated and forwarded directly based on the flow affinity table attributes.

[0212] S6: When the computing power routing and forwarding component of the ingress gateway generates the flow affinity table, it also sends the OAM detection enable information of the corresponding flow session quality to the computing network OAM component. The elements of the OAM detection enable information include: the source IP of the flow affinity table, computing power identifier, detection cycle, and SLA expected indicators (packet loss, latency, connectivity).

[0213] S7: The computing network OAM component of the ingress gateway constructs an OAM message based on the enabling information and sends it to the computing network OAM component. The encapsulation format of the OAM message is determined by the selected OAM protocol. Its destination IP is the computing power identifier and the source IP is the source IP of the flow affinity table. In addition, it should be pointed out that you can choose to use in-flow detection such as in-situ network telemetry (INT), in-situ operation administration and maintenance (IOAM), in-situ flow analyzer (IFA) and other technologies to detect business flow performance. You can also choose to directly detect the round-trip time (RTT) of the business flow based on the chip flow table.

[0214] S8: The computing routing and forwarding component of the ingress gateway searches the flow affinity table for the SR-Policy, egress gateway, and computing instance information based on the destination and source addresses of the OAM message. It then encapsulates the tunnel header and SRH information and forwards the message.

[0215] S9-S11: The computing power routing forwarding component of the egress gateway decapsulates the tunnel header and SRH and forwards the OAM message to the compute instance based on the compute instance IP address. Based on the deployed OAM service and the OAM message protocol, the compute instance responds with information such as packet loss and latency. The destination IP address of the response message is the source IP address of the OAM message, and the source IP address of the response message is the compute instance IP address. The computing power routing forwarding component of the egress gateway encapsulates the received compute instance OAM response message in the downlink tunnel and forwards it.

[0216] S12: The computing routing and forwarding component of the ingress gateway receives the OAM response message, identifies and extracts the OAM response message based on the message type and destination IP address, and sends it to the computing network OAM component. This corresponds to Figure 10. Computing network OAM enables user-oriented service flow assessment of service-level computing network end-to-end service-level agreement (E2E SLA) targets (including but not limited to computing network packet loss, computing network latency, and other computing network metrics).

[0217] S13: The computing network OAM component compares the detection result data with the expected SLA indicators. If the indicators are lower than the expected SLA, it sends a path compensation calculation request to the computing power routing calculation component. The request elements include: the terminal source IP address, the computing power identifier as the key value, the network side SR-Policy SLA requirement difference, and the head node and tail node IP addresses.

[0218] S14: The computing power routing calculation component of the ingress gateway calculates new SR-Policy requirement parameters based on the path compensation requirements sent by the computing network OAM component. The new requirement parameters are the original SR-Policy SLA parameters superimposed with the compensation requirement parameters. SR-TE is constrained by the new requirement parameters to calculate a new SR-Policy. This is sent to the forwarding plane along with the terminal source IP address and computing power identification key value to refresh the flow affinity table, providing a new path service to meet the end-to-end computing network service access requirements while the service instance remains unchanged.

[0219] The embodiment of the present application is based on a computing network (including a service-aware network) for IPv6 / SRv6. The proposed computing network OAM architecture can achieve rapid convergence of computing power routing and reliable quality assurance of computing network services. The computing network OAM architecture aims to establish a computing network business closed-loop system based on computing network OAM. The slow lane control plane slowly collects / measures + calculates to form a business forwarding plan, and completes the basic business action control of the business through the issuance of rules. The fast lane computing network OAM component quickly collects / measures + calculates to make corrections / delimitations, and performs business optimization through control. The entire computing network system is a negative feedback closed-loop system that can provide a better computing network service with a consistent experience. The computing network OAM component is introduced into the computing network business closed-loop system to realize fault and performance detection from the network domain to the computing domain. The detection scope covers the computing instance connected to the egress gateway, the computing power routing table SLA of the ingress gateway, and the business flow SLA of the ingress gateway. The detection objects of the computing network OAM component include three types of computing instances, computing power routing tables, and business flows. In addition, the component also includes a computing network OAM management and execution module. The computing network OAM management module is responsible for configuring and enabling objects issued by the OAM application component. It initiates OAM detection instances for different configuration objects and demand levels, receives results from the OAM execution component, and publishes detection results according to the object-enabled subscription demand side (i.e., the OAM application component) to trigger computing network actions. The computing network OAM execution module executes OAM instructions on the detection objects, encapsulates and sends detection messages and configurations to the computing power routing and forwarding component, receives OAM response messages from the computing power routing and forwarding component, and organizes detection results to trigger the reporting management component based on thresholds. To accelerate the forwarding plane, hardware including FPGAs and dedicated NPs is implemented to achieve separation between the execution fast lane and the management slow lane. The decoupling of computing network OAM components and detection protocols enables robust detection protocol expansion. Computing network OAM detection protocols may include, but are not limited to, the Two-Way Active Measurement Protocol (TWAMP), the One-Way Active Measurement Protocol (OWAMP), the Bidirectional Forwarding Detection Protocol (BFD), or combinations thereof or detection capability extensions. New detection protocols can also be constructed to detect metrics between the network and the compute instance IP, including but not limited to connectivity and packet loss, latency, and bandwidth. The computing network OAM components are deployed on the egress gateway, with the connected compute instance IP as the destination address to initiate compute instance-level OAM. FM, PM, and XM-related metrics are detected from the egress gateway to the compute instance. Based on the detection results, the egress gateway's effective computing power status is triggered to propagate, accelerating the rapid convergence of control plane computing power routing.The computing network OAM component is deployed at the ingress gateway. It initiates computing network routing-level OAM for each entry in the computing network routing table, carrying different ServiceIDs. It checks the FM, PM, and XM metrics of the current service routing table from the ingress gateway to the compute instance, verifying whether the computing network SLA provided by the latest computing network routing table is met. For entries that do not meet the requirements, a rapid convergence update of the control plane is triggered. The computing network OAM component is deployed at the ingress gateway and targets the flow affinity table. Two detection models are supported for flows that match the flow affinity table: Method 1: If the service host supports in-flow detection, in-flow detection is enabled directly for the service flow. Method 2: Probe packets are directly constructed on the data plane to implement service flow-level OAM. Both methods detect the FM, PM, and XM metrics of the actual forwarded service flow. Based on the detection results, the user service flow is evaluated for achieving the computing network SLA goals. Since computing network access requires flow affinity, if end-to-end service flow access metrics are detected to be degraded, the network path is dynamically adjusted to repair and guarantee the end-to-end service flow metrics, ensuring a consistent user experience.

[0220] Figure 11 is a structural diagram of an electronic device provided in an embodiment of the present application, which electronic device includes a processor 10 and a memory 11; the number of processors 10 in the electronic device can be one or more, and Figure 11 takes one processor 10 as an example; the processor 10 and the memory 11 in the electronic device can be connected via a bus or other means, and Figure 11 takes connection via a bus as an example.

[0221] Memory 11, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the components corresponding to the computing network operation, maintenance, and management system in the embodiments of the present application. Processor 10 executes the software programs, instructions, and modules stored in memory 11 to execute various functional applications and data processing of the electronic device, thereby implementing the aforementioned computing network operation, maintenance, and management method.

[0222] The memory 11 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created based on the use of the electronic device, etc. In addition, the memory 11 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 11 may further include a memory remotely located relative to the processor 10, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0223] The present application also provides a storage medium containing computer-executable instructions. When the computer-executable instructions are executed by a computer processor, the computer-executable instructions are used to perform a method for operating, maintaining, and managing a computing network. The method includes:

[0224] The computing power perception component obtains the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and sends an enabling message to the computing network operation and maintenance management component based on the computing power information to start the computing network operation and maintenance management component;

[0225] The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0226] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, so that the computing instance generates an operation and maintenance management message response according to the deployed operation and maintenance management service and the operation and maintenance management message protocol;

[0227] The computing power routing and forwarding component receives the operation and maintenance management message response of the computing instance, and sends the operation and maintenance management message response to the computing network operation and maintenance management component;

[0228] The computing network operation and maintenance management component generates an operation and maintenance management detection result according to the operation and maintenance management message response, and determines an operation and maintenance management detection notification message according to the detection object information;

[0229] The computing power perception component receives the operation maintenance management detection notification message and spreads the computing power information corresponding to the operation maintenance management detection notification message;

[0230] The upstream node receives and processes the computing power information, recalculates the computing power routing and sends it to the computing power routing forwarding component.

[0231] and / or,

[0232] The computer executable instructions, when executed by a computer processor, are used to perform a method for operating, maintaining and managing a computing network, the method comprising:

[0233] The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component;

[0234] The computing power routing calculation component generates a computing power routing table according to the measurement process information, and sends the computing power routing table to the computing power routing forwarding component;

[0235] The computing power routing calculation component sends enabling information to the computing network operation and maintenance management component according to the computing power routing table;

[0236] The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0237] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component;

[0238] The computing network operation and maintenance management component generates a detection result of the computing power routing table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component;

[0239] The computing power routing calculation component determines a service level agreement target gap based on the detection result, recalculates a computing power routing table based on the service level agreement target gap, and sends the computing power routing table to the computing power routing forwarding component.

[0240] and / or,

[0241] The computer executable instructions, when executed by a computer processor, are used to perform a method for operating, maintaining and managing a computing network, the method comprising:

[0242] The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component;

[0243] The computing power routing calculation component generates a computing power routing table according to the measurement process information, and sends the computing power routing table to the computing power routing forwarding component;

[0244] The computing power routing forwarding component determines the flow affinity table based on the received computing power routing table and the computing power service access request of the terminal user, and sends the enabling information to the computing network operation and maintenance management component;

[0245] The computing network operation and maintenance management component constructs an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component;

[0246] The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component;

[0247] The computing network operation and maintenance management component determines a detection result based on the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component;

[0248] The computing power routing calculation component determines the service level agreement target gap corresponding to the detection result, calculates the path compensation according to the service level agreement target gap, and sends the path compensation to the computing power routing forwarding component to refresh the flow affinity table.

[0249] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present application can be implemented with the help of software and necessary general-purpose hardware, and of course it can also be implemented with hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the carrier aggregation method described in each embodiment of the present application.

[0250] It is worth noting that in the embodiment of the above-mentioned device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application.

[0251] Those skilled in the art will appreciate that all or some of the steps, devices, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0252] In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. The corresponding software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or temporary media). As is well known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media includes, but is not limited to, RAM, ROM, Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory or other memory technology, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0253] The above content illustrates the optional embodiments of the present application with reference to the accompanying drawings, and does not limit the scope of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present application should be within the scope of the present application.

Claims

1. A computing network operation and maintenance management system, applied to computing power gateway devices, comprising: Network perception component, computing power perception component, computing power routing calculation component, computing power routing forwarding component, and computing network operation, maintenance and management component; The network awareness component is used for awareness and diffusion of network information of network nodes and links; The computing power perception component is used to connect to the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and to receive and disseminate the computing power information of the computing power gateway device; The computing power routing calculation component is connected to the network perception component and the computing power perception component respectively, and is used to summarize the measurement indicator information corresponding to the network and computing power respectively, and dynamically determine the computing power routing according to the measurement indicator information; The computing power routing forwarding component is used to receive the computing power routing determined by the computing power routing calculation component, and provide computing network integrated forwarding for the service message according to the computing power routing; The computing network operation and maintenance management component performs operation and maintenance management detection based on the objects received from the computing power perception component and the computing power routing calculation component and reports the detection results.

2. The system according to claim 1, wherein: The computing power perception component is used to connect to the computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and to notify the computing power information to the protocol diffusion, including: The computing power perception component is deployed on the export computing power gateway device or network controller, and is used to receive the computing power information collected by the cloud-side computing power collection component and notify the protocol diffusion of the computing power information.

3. The system according to claim 1, wherein: The system collects computing network status based on the cloud-side computing power collection component, the computing power perception component, and the network perception component. The computing network status collection at least includes: The cloud-side computing power collection component notifies the computing power perception component of the collected computing power information to notify the protocol to perform network domain diffusion to form a computing power status database; The network domain collects network status information through the network perception component to form a network status database.

4. The system according to claim 1, wherein: The system performs computing network joint detection and analysis based on the computing power routing calculation component and the computing power routing forwarding component, wherein the computing network joint detection and analysis includes at least one of the following: The computing power routing calculation component determines a computing power routing table based on the computing power measurement index information and the network measurement index information, and sends the computing power routing table to the computing power routing forwarding component. In addition, the computing power routing calculation component sends enabling information to the computing network operation and maintenance management component based on the computing power routing table to trigger the computing network operation and maintenance management component to perform detection and feedback the detection results; The computing power routing forwarding component generates a flow affinity table based on the computing power routing table and the computing power service access request of the terminal user, and the computing power routing forwarding component sends an enabling message to the computing network operation and maintenance management component to trigger the computing network operation and maintenance management component to perform detection and feedback the detection result; The computing power routing calculation component performs a service level agreement target gap analysis based on the detection results.

5. The method according to claim 4, further comprising: The computing power routing calculation component receives the flow affinity table and manages and maintains the user session according to the flow affinity table.

6. The system according to claim 1 further comprising: The system measures the detection object according to the computing network operation and maintenance management component and the computing power routing and forwarding component, wherein the detection object includes: at least one of: a computing instance, a computing power routing table, and a flow corresponding to a flow affinity table.

7. The system according to claim 6, wherein: The detection object includes a computing instance, and the system measures the detection object according to the computing network operation and maintenance management component and the computing power routing and forwarding component, including: The computing network operation and maintenance management component receives the enabling information sent by the computing power perception component based on the computing power information, assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component; The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component; The computing network operation and maintenance management component detects the computing instance according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component.

8. The system according to claim 6, wherein: The detection object includes a computing power routing table, and the system measures the detection object according to the computing network operation and maintenance management component and the computing power routing forwarding component, including: The computing power routing forwarding component receives the computing power routing table determined by the computing power routing calculation component according to the computing power measurement index information and the network measurement index information; The computing network operation and maintenance management component receives the enabling information issued by the computing power routing calculation component according to the computing power routing entry of the computing power routing table, assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing forwarding component; The computing power routing forwarding component deployed on the ingress gateway encapsulates and forwards the operation maintenance management message to the computing power routing forwarding component deployed on the egress gateway; The computing power routing forwarding component deployed on the egress gateway decapsulates the operation and maintenance management message, sends the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, encapsulates the operation and maintenance management message response and forwards it to the computing power routing forwarding component deployed on the ingress gateway; The computing power routing forwarding component deployed on the ingress gateway decapsulates the operation maintenance management message response and sends the operation maintenance management message response to the computing network operation maintenance management component; The computing network operation and maintenance management component detects the computing power routing table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component.

9. The system according to claim 6, wherein: The detection object includes a flow corresponding to a flow affinity table, and the system measures the detection object according to the computing network operation and maintenance management component and the computing power routing and forwarding component, including: The computing power routing forwarding component receives the computing power routing table determined by the computing power routing calculation component according to the computing power measurement index information and the network measurement index information; The computing power routing forwarding component deployed on the ingress gateway receives the computing power service access request from the terminal user, determines the flow affinity table based on the computing power service access request and the computing power routing table, and sends enabling information to the computing network operation and maintenance management component; The computing network operation and maintenance management component constructs an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component deployed on the ingress gateway; The computing power routing forwarding component deployed on the ingress gateway encapsulates and forwards the operation maintenance management message to the computing power routing forwarding component deployed on the egress gateway; The computing power routing forwarding component deployed on the egress gateway decapsulates the operation and maintenance management message, sends the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, encapsulates the operation and maintenance management message response and forwards it to the computing power routing forwarding component deployed on the ingress gateway; The computing power routing forwarding component deployed on the ingress gateway decapsulates the operation maintenance management message response and sends the operation maintenance management message response to the computing network operation maintenance management component; The computing network operation and maintenance management component detects the flow affinity table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component.

10. The system according to claim 1, wherein: The system performs network operation, maintenance, and management measurement feedback based on the computing power perception component and the computing power routing calculation component, wherein the network operation, maintenance, and management measurement feedback includes at least one of the following: The computing power routing calculation component receives the detection result, recalculates and corrects the detection object corresponding to the detection result; The computing power sensing component receives the detection result and notifies the protocol diffusion of the detection result.

11. The system according to claim 10, wherein: The detection object includes at least one of the flows corresponding to the computing power routing table and the flow affinity table.

12. The system according to claim 1, wherein: The computing network operation and maintenance management component includes at least one of the following: a computing network operation and maintenance management configuration module and a computing network operation and maintenance management execution module.

13. A computing network operation and maintenance management method, applied to a computing network operation and maintenance management system, the method comprising: The computing power perception component obtains computing power information collected by the cloud-side computing power collection component deployed in the cloud resource pool, and sends enabling information to the computing network operation and maintenance management component based on the computing power information to start the computing network operation and maintenance management component; The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component; The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, so that the computing instance generates an operation and maintenance management message response according to the deployed operation and maintenance management service and the operation and maintenance management message protocol; The computing power routing and forwarding component receives the operation and maintenance management message response of the computing instance, and sends the operation and maintenance management message response to the computing network operation and maintenance management component; The computing network operation and maintenance management component generates an operation and maintenance management detection result according to the operation and maintenance management message response, and determines an operation and maintenance management detection notification message according to the detection object information; The computing power perception component receives the operation maintenance management detection notification message and spreads the computing power information corresponding to the operation maintenance management detection notification message; The upstream node receives and processes the computing power information, recalculates the computing power routing and sends it to the computing power routing forwarding component.

14. The method according to claim 13, further comprising: The computing network operation and maintenance management component determines a detection object according to the operation and maintenance management detection result, and determines a feedback notification according to a configuration threshold of the detection object and the operation and maintenance management detection result.

15. A computing network operation and maintenance management method, applied to a computing network operation and maintenance management system, the method comprising: The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component; The computing power routing calculation component generates a computing power routing table based on the measurement process information and sends the Computing power routing table to computing power routing forwarding component; The computing power routing calculation component sends enabling information to the computing network operation and maintenance management component according to the computing power routing table; The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component; The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component; The computing network operation and maintenance management component generates a detection result of the computing power routing table according to the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component; The computing power routing calculation component determines a service level agreement target gap based on the detection result, recalculates a computing power routing table based on the service level agreement target gap, and sends the computing power routing table to the computing power routing forwarding component.

16. The method according to claim 15, wherein: The computing network operation and maintenance management component assembles an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component, including: The computing network operation and maintenance management component assembles the operation and maintenance management message according to the enabling information, wherein the message format of the operation and maintenance management message is determined according to the operation and maintenance management protocol selected by the computing network operation and maintenance management component; The computing network operation and maintenance management component sends the operation and maintenance management message to the computing power routing and forwarding component deployed on the ingress gateway.

17. The method according to claim 15, wherein: The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component, including: The computing power routing forwarding component deployed on the ingress gateway encapsulates the received operation maintenance management message and forwards the operation maintenance management message to the computing power routing forwarding component deployed on the egress gateway; The computing power routing forwarding component deployed on the egress gateway decapsulates the operation and maintenance management message, sends the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, encapsulates the operation and maintenance management message response and forwards it to the computing power routing forwarding component deployed on the ingress gateway; The computing power routing forwarding component deployed on the ingress gateway decapsulates the operation and maintenance management message response and sends the operation and maintenance management message response to the computing network operation and maintenance management component.

18. A computing network operation and maintenance management method, applied to a computing network operation and maintenance management system, the method comprising: The network perception component and the computing power perception component respectively send measurement process information to the computing power routing calculation component; The computing power routing calculation component generates a computing power routing table according to the measurement process information, and sends the computing power routing table to the computing power routing forwarding component; The computing power routing forwarding component determines the flow affinity table based on the received computing power routing table and the computing power service access request of the terminal user, and sends the enabling information to the computing network operation and maintenance management component; The computing network operation and maintenance management component constructs an operation and maintenance management message according to the enabling information, and sends the operation and maintenance management message to the computing power routing and forwarding component; The computing power routing forwarding component forwards the operation and maintenance management message to the computing instance, receives the operation and maintenance management message response fed back by the computing instance, and forwards the operation and maintenance management message response to the computing network operation and maintenance management component; The computing network operation and maintenance management component determines a detection result based on the operation and maintenance management message response, and feeds back the detection result to the computing power routing calculation component; The computing power routing calculation component determines the service level agreement target gap corresponding to the detection result, calculates the path compensation according to the service level agreement target gap, and sends the path compensation to the computing power routing forwarding component to refresh the flow affinity table.

19. The method according to claim 18, wherein: The computing power routing calculation component determines a service level agreement target gap corresponding to the detection result, calculates a path compensation based on the service level agreement target gap, and sends the path compensation to the computing power routing forwarding component to refresh the flow affinity table, including: The computing power routing calculation component determines a detection result of the flow corresponding to the flow affinity table, and if the detection result is lower than the expected service level agreement target of the flow corresponding to the flow affinity table, generates the path compensation corresponding to the service level agreement target gap, and sends the path compensation to the computing power routing table to refresh the flow affinity table.

20. A computer-readable storage medium storing one or more programs, wherein the one or more programs are executed by one or more processors to implement the computing network operation, maintenance and management method according to any one of claims 13 to 19.

Images