Computing power access method, apparatus, network device and storage medium

Abstract

Computing power access methods, an apparatus, a network device, and a storage medium, relating to the technical field of communications. A method comprises: a first gateway device receiving a first computing power access message sent by a client, the client accessing a computing power network by means of the first gateway device, and the destination address of the first computing power access message being the address of a target computing power service; on the basis of a pre-stored first computing power routing table, determining the address of a second gateway device corresponding to the address of the target computing power service, the first computing power routing table containing addresses of computing power services and addresses of gateway devices to which service nodes where service instances providing the computing power services are located are connected when accessing the computing power network; on the basis of the address of the first gateway device and the address of the second gateway device, encapsulating the first computing power access message to obtain a second computing power access message; and sending the second computing power access message to the second gateway device. This application provides a data plane forwarding technology, which implements computing power access.

Description

A computing power access method, device, network device and storage medium Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a computing power access method, apparatus, network equipment, and storage medium. Background Art

[0002] The computing power network is a new type of information infrastructure that provides integrated services to customers by sensing the network status and the computing power status of the cloud. Currently, the computing power network uses a bearer network based on IP Version 6 (IPv6) segment routing (SRv6) technology as the data plane. Clients access the server through the SRv6 path to obtain computing power services from the server. However, the method of using SRv6 technology for data plane forwarding to achieve computing power access is only a concept, and there is no confirmed data plane forwarding technology to achieve computing power access to the computing power network.

[0003] Summary of the Invention

[0004] The purpose of the embodiments of this application is to provide a computing power access method, apparatus, network device, and storage medium to provide a data plane forwarding technology to implement computing power access. The specific technical solution is as follows:

[0005] In a first aspect, an embodiment of the present application provides a computing power access method, applied to a first gateway device, the method comprising:

[0006] Receive a first computing power access message sent by a client, the client accessing the computing power network through the first gateway device, and the destination address of the first computing power access message is the address of the target computing power service;

[0007] Determining, based on a pre-stored first computing power routing table, an address of a second gateway device corresponding to the address of the target computing power service, wherein the first computing power routing table includes the address of the computing power service and the address of a gateway device to which a service node where a service instance providing the computing power service is located is connected when the service node accesses the computing power network;

[0008] Encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain a second computing power access message;

[0009] Send the second computing power access message to the second gateway device.

[0010] In a second aspect, an embodiment of the present application provides a computing power access method, applied to a second gateway device, the method comprising:

[0011] Receive a second computing power access message from a first gateway device, where the first gateway device is the gateway device connected when the client accesses the computing power network. The second computing power access message is a message obtained by encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device after the first gateway device determines the address of the second gateway device based on a pre-stored first computing power routing table and the destination address of the first computing power access message sent by the client. The destination address of the first computing power access message is the address of the target computing power service. The first computing power routing table includes the address of the computing power service and the address of the gateway device connected when the service node where the service instance providing the computing power service is located accesses the computing power network.

[0012] Decapsulating the second computing power access message to obtain a third computing power access message;

[0013] Determining, based on a pre-stored second computing power routing table, an address of a target proxy device corresponding to the address of the target computing power service, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service;

[0014] According to the address of the target proxy device, the third computing power access message is sent to the target service instance that provides the target computing power service.

[0015] In a third aspect, an embodiment of the present application provides a computing power access method, which is applied to an operation and maintenance platform. The method includes:

[0016] Get the name of the target computing service from the client;

[0017] Determine the address of the target computing service corresponding to the name of the target computing service according to the pre-stored correspondence between the name of the computing service and the address of the computing service;

[0018] The address of the target computing service is sent to the client, so that the client can access the target service node where the target service instance providing the target computing service is located through the first gateway device and the second gateway device in the computing network according to the address of the target computing service. The client accesses the computing network through the first gateway device, and the target service node accesses the computing network through the second gateway device.

[0019] In a fourth aspect, an embodiment of the present application provides a computing power access method, which is applied to a client, wherein the client accesses a computing power network through a first gateway device, and the method includes:

[0020] Get the address of the target computing power service;

[0021] Generate a first computing power access message with a destination address being an address of the target computing power service;

[0022] A first computing power access message is sent to the first gateway device, and the client accesses the computing power network through the first gateway device, so that the first gateway device sends the first computing power access message to the target service node where the target service instance providing the target computing power service is located through the second gateway device, and the target service node accesses the computing power network through the second gateway device.

[0023] In a fifth aspect, an embodiment of the present application provides a computing power access device, applied to a first gateway device, the device comprising:

[0024] A first receiving module is configured to receive a first computing power access message sent by a client, wherein the client accesses the computing power network through the first gateway device, and the destination address of the first computing power access message is the address of a target computing power service;

[0025] A determination module is configured to determine an address of a second gateway device corresponding to the address of the target computing service based on a pre-stored first computing power routing table, wherein the first computing power routing table includes the address of the computing power service and the address of the gateway device to which the service node where the service instance providing the computing power service is located is connected when accessing the computing power network;

[0026] an encapsulation module, configured to encapsulate the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain a second computing power access message;

[0027] A sending module is used to send the second computing power access message to the second gateway device.

[0028] In a sixth aspect, an embodiment of the present application provides a computing power access device, applied to a second gateway device, the device comprising:

[0029] A first receiving module is configured to receive a second computing power access message from a first gateway device, where the first gateway device is a gateway device to which a client is connected when accessing a computing power network. The second computing power access message is a message obtained by encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device after the first gateway device determines the address of the second gateway device based on a pre-stored first computing power routing table and the destination address of the first computing power access message sent by the client. The destination address of the first computing power access message is the address of a target computing power service. The first computing power routing table includes the address of the computing power service and the address of the gateway device to which a service node where a service instance providing the computing power service is located is connected when accessing the computing power network.

[0030] a decapsulation module, configured to decapsulate the second computing power access message to obtain a third computing power access message;

[0031] a determination module, configured to determine an address of a target proxy device corresponding to the address of the target computing service based on a pre-stored second computing power routing table, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service;

[0032] The first sending module is used to send the third computing power access message to the target service node where the target service instance providing the target computing power service is located according to the address of the target proxy device.

[0033] In a seventh aspect, an embodiment of the present application provides a computing power access device, applied to an operation and maintenance platform, the device comprising:

[0034] The first acquisition module is used to obtain the name of the target computing service from the client;

[0035] A first determining module is configured to determine the address of a target computing service corresponding to the name of the target computing service based on a pre-stored correspondence between the name of the computing service and the address of the computing service;

[0036] The first sending module is used to send the address of the target computing power service to the client, so that the client can access the target service node where the target service instance providing the target computing power service is located through the first gateway device and the second gateway device in the computing power network according to the address of the target computing power service. The client accesses the computing power network through the first gateway device, and the target service node accesses the computing power network through the second gateway device.

[0037] In an eighth aspect, an embodiment of the present application provides a computing power access device, which is applied to a client, wherein the client accesses a computing power network through a first gateway device, and the device includes:

[0038] The acquisition module is used to obtain the address of the target computing power service;

[0039] A generating module, configured to generate a first computing power access message whose destination address is an address of the target computing power service;

[0040] A sending module is used to send a first computing power access message to the first gateway device, and the client accesses the computing power network through the first gateway device, so that the first gateway device sends the first computing power access message to the target service node where the target service instance providing the target computing power service is located through the second gateway device, and the target service node accesses the computing power network through the second gateway device.

[0041] In the ninth aspect, an embodiment of the present application provides a network device, comprising a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions that can be executed by the processor, and the processor is prompted by the machine-executable instructions to: implement any method provided in the first aspect, or implement any method provided in the second aspect, or implement any method provided in the third aspect, or implement any method provided in the fourth aspect.

[0042] In the tenth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements any method provided in the first aspect, or any method provided in the second aspect, or any method provided in the third aspect, or any method provided in the fourth aspect.

[0043] In the eleventh aspect, an embodiment of the present application provides a computer program product, which, when running on a computer, enables the computer to execute any method provided in the first aspect, or execute any method provided in the second aspect, or execute any method provided in the third aspect, or execute any method provided in the fourth aspect.

[0044] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device in the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0045] Of course, it is not necessary to achieve all the advantages described above at the same time when implementing any product or method of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0047] FIG1 is a schematic diagram of a centralized computing network in related art;

[0048] FIG2 is a schematic diagram of a distributed computing network in related art;

[0049] FIG3 is a schematic diagram of a computing power access system provided in an embodiment of the present application;

[0050] FIG4 is a schematic diagram of a first flow chart of a computing power access method provided in an embodiment of the present application;

[0051] FIG5 is a schematic diagram of a CSID provided in an embodiment of the present application;

[0052] FIG6 is a schematic diagram of a first computing power access message provided in an embodiment of the present application;

[0053] FIG7 is a schematic diagram of a second flow chart of a computing power access method provided in an embodiment of the present application;

[0054] FIG8a is a first schematic diagram of a second computing power access message provided in an embodiment of the present application;

[0055] FIG8 b is a second schematic diagram of a second computing power access message provided in an embodiment of the present application;

[0056] FIG8c is a third schematic diagram of a second computing power access message provided in an embodiment of the present application;

[0057] FIG9 is a schematic diagram of a third flow chart of the computing power access method provided in an embodiment of the present application;

[0058] FIG10 is a flow chart of a method for allocating addresses for computing power access provided in an embodiment of the present application;

[0059] FIG11 is a schematic diagram of a fourth flow chart of the computing power access method provided in an embodiment of the present application;

[0060] FIG12 is a signaling diagram of the preparation phase for computing power access according to an embodiment of the present application;

[0061] FIG13 is a signaling diagram of the execution phase of computing power access provided by an embodiment of the present application;

[0062] FIG14 is a schematic diagram of the first structure of a computing power access device provided in an embodiment of the present application;

[0063] FIG15 is a schematic diagram of a second structure of a computing power access device provided in an embodiment of the present application;

[0064] FIG16 is a schematic diagram of a third structure of a computing power access device provided in an embodiment of the present application;

[0065] FIG17 is a schematic diagram of a fourth structure of a computing power access device provided in an embodiment of the present application;

[0066] FIG18 is a schematic diagram of the structure of a network device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] To make the objectives, technical solutions, and advantages of this application more clearly understood, the present application is further described below with reference to the accompanying drawings and examples. It is apparent that the described examples are only a portion of the embodiments of this application, and not all of them. All other embodiments derived by persons of ordinary skill in the art based on the examples in this application are intended to fall within the scope of protection of this application.

[0068] To facilitate understanding, the terms appearing in the embodiments of this application are explained below.

[0069] Computing resource pool: A resource pool that provides computing resources or computing services. Computing resource pools can exist in cloud form, including central clouds, regional clouds, and edge clouds. Computing resource pools can also exist in non-cloud form, with no restrictions.

[0070] Computing power network: Computing power resource pools at different levels, such as cloud, edge, and end, are interconnected through a high-speed network to form a large-scale system that integrates computing power and network. This system can provide unified social-level computing power services to the outside world. Among them, cloud is short for computing cloud, which is a supercomputing model based on the network "cloud". In remote data centers, thousands of computers and servers are connected to form a computing cloud; edge is short for multi-access edge computing (MEC) network architecture, which provides cloud computing functions and information technology (IT) service environment at the edge of the network; end is short for terminal, which refers to all devices with internet and computing capabilities.

[0071] Computing Service Identity (CSID): A logical concept used to identify a computing service that is deployed in a computing resource pool and can be provided to the outside world. The computing service indicated by the CSID has nothing to do with the physical location of the entity that actually provides the computing service. A computing service can be composed of multiple computing service instances (CSIs) with the same functions. These CSIs can be located in computing resource pools at different physical locations. In the embodiment of the present application, the CSID can represent the address of the computing service, and the computing service instance is referred to as the service instance.

[0072] Computing Service Instance Identity (CSIID): used to identify a service instance that actually provides computing power services under a certain CSID, such as the above-mentioned CSI. In an embodiment of the present application, the service instance that provides computing power services is deployed on a service node, and one or more service instances can be deployed on a service node. In an embodiment of the present application, CSIID can represent the address of a service instance. For multiple service instances deployed on the same service node, the CSIIDs of these multiple service instances may be different or the same, such as the CSIIDs of multiple service instances are all the IP addresses of the service nodes where they are located. However, for service instances with the same CSIID, the computing power services they provide are different, that is, their corresponding CSIDs are different.

[0073] Client: The initiator of computing power request.

[0074] Server: The service provider for computing power requests. The server can be located on a cloud server or a non-cloud server, such as the aforementioned central cloud, regional cloud, and edge cloud computing resource pools. The server can include one or more service nodes.

[0075] Gateway (GW): An edge device in a computing network. GWs include ingress GWs and egress GWs. In this embodiment, GWs support computing routing.

[0076] The computing power network is a new type of information infrastructure. It provides integrated services to clients by simultaneously sensing the network status and the computing power status of the cloud.

[0077] The computing network includes an operation and maintenance platform, a network controller, a cloud management platform, a computing network, and a server. The operation and maintenance platform is the brain of the computing network, referred to as the computing network brain, a centralized upper-layer software platform. The operation and maintenance platform is used for the orchestration and scheduling of computing services, and for centralized control of computing resources in the computing network. The network controller is a centralized upper-layer software platform used to issue network policies and paths. The cloud management platform is a software platform for managing and monitoring computing resource pools. Provider Edge (PE) devices and intermediate devices are deployed in the computing network. The PE device is a GW. The PE device connected to the client can be called a network PE, that is, an ingress GW, and the PE device connected to the server can be called a cloud PE, that is, an egress GW. Intermediate devices are used for routing and forwarding between PE devices.

[0078] In terms of technical route, computing power networks have two forms: centralized computing power networks and distributed computing power networks.

[0079] Figure 1 shows the architecture of a centralized computing network. In a centralized computing network, the operations and maintenance platform orchestrates and schedules computing services, providing centralized control over computing resources within the network. The operations and maintenance platform collects computing and network information from the entire computing network. Based on this information, it maintains the computing and network topologies of the computing network, selects appropriate clouds for clients, and establishes network paths.

[0080] The architecture of the distributed computing network can be seen in Figure 2. In the distributed computing network, computing network information is transmitted in the computing network, so that all PE devices maintain computing network information such as computing power information and network information. Each PE device can perform strategic routing based on the computing network information to provide appropriate computing power services to the client.

[0081] Currently, the computing power network uses a bearer network based on Segment Routing over IPv6 (SRv6) technology as the data plane. Clients access the server through the SRv6 path to obtain computing power services from the server. However, using SRv6 technology for data plane forwarding to achieve computing power access is only a concept, and there is no confirmed data plane forwarding technology to achieve computing power access on the computing power network.

[0082] To implement computing power access on a computing power network, an embodiment of the present application provides a computing power access system. As shown in Figure 3, the computing power access system is divided into a management and control layer 31 and a device layer 32. The management and control layer 31 may include an operation and maintenance platform 311, a Domain Name System (DNS) server 312, and a network controller 313. The device layer 32 may include a client 321, a gateway device 322, a gateway device 323, a proxy device 324, and a server 325.

[0083] The operation and maintenance platform 311 is used to orchestrate and schedule the computing services provided by the computing network. The operation and maintenance platform 311 can also be called the brain of the computing network.

[0084] The DNS server 312 is used to provide domain name resolution services and is in communication with the operation and maintenance platform 311 .

[0085] The network controller 313 is used to issue network policies and paths. The network controller 313 is in communication with the operation and maintenance platform 311. After the operation and maintenance platform 311 formulates the network policies and paths, it issues them to gateway devices, such as gateway device 323 in Figure 3, through the network controller 313. In the embodiments of this application, the communication between the operation and maintenance platform 311 and the gateway devices is achieved through the network controller 313. For ease of description, the network controller 313 is omitted in the following description.

[0086] In the embodiment of the present application, the operation and maintenance platform 311, the DNS server 312 and the network controller 313 can be deployed on different physical machines or integrated on the same physical machine, which is not limited.

[0087] The client 321 is used to initiate a computing power access request to request computing power services, such as sending a computing power access message to the gateway device 322. The client 321 can be a mobile terminal, a personal computer (PC), a smart home or enterprise terminal, etc.

[0088] The server 325 is used to provide computing power services, such as receiving computing power access messages and processing the computing power access messages. The server 325 can be a computing power resource pool such as a central cloud, edge cloud, or regional cloud.

[0089] In a computing network, a server 325 may include one or more service instances. A service instance provides a computing service, such as a computing service for image processing. When a server 325 includes multiple service instances, it can provide one or more computing services to the outside world.

[0090] Furthermore, a computing service can be implemented by a single service instance in a server 325, or by multiple service instances in one or more servers 325. For example, computing service 1 for image processing is implemented by service instance a1 in server a and service instance b1 in server b, and computing service 2 for image processing is implemented by service instance a2 in server a and service instances b2-b3 in server b.

[0091] The computing network includes multiple routing nodes, such as gateway device 322, gateway device 323, and other intermediate nodes. Gateway device 322 is the entry gateway device of the computing network, providing access to the computing network for client 321. Gateway device 323 is the exit gateway device of the computing network, providing access to the computing network for server 325.

[0092] Proxy device 324 is a proxy for the computing resource pool, specifically a proxy for server 325. It can also be referred to as a proxy for a service node within server 325 or a proxy for a service instance within server 325. Proxy device 324 is used to collect computing network information, including static and dynamic attribute information, of service instances within server 325. Static attribute information may include one or more of the following: the address of the computing service within server 325, the name of the computing service, the service type of the computing service, the address of the service instance providing the computing service, the location of the service instance, the priority of the service instance, and the capabilities of the service instance. Dynamic attribute information may include one or more of the load status and session status of the service instance.

[0093] The proxy device 324 can be located within the server 325, such as when the proxy device 324 and a service instance are located on the same service node. The proxy device 324 can also be located outside the server 325, such as when the proxy device 324 is a separate physical machine. The embodiment of the present application does not specifically limit the location of the proxy device 324.

[0094] Based on the above computing power access system, an embodiment of the present application provides a computing power access method. As shown in Figure 4, this computing power access method is applied to a client, which accesses the computing power network through a first gateway device. The first gateway device can be any gateway device in the computing power network. The computing power access method includes the following steps.

[0095] Step S41, obtaining the address of the target computing power service;

[0096] Step S42: Generate a first computing power access message whose destination address is the address of the target computing power service;

[0097] Step S43: Send a first computing power access message to the first gateway device. The client accesses the computing power network through the first gateway device, so that the first gateway device sends the first computing power access message to the target service node where the target service instance providing the target computing power service is located through the second gateway device. The target service node accesses the computing power network through the second gateway device.

[0098] In the technical solution provided in the embodiment of the present application, an address is assigned to the computing power service, the client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device in the computing power network. Then, the first gateway device and the second gateway device in the computing power network query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node. The target service node uses the target service instance to process the computing power access message, realizes data plane forwarding, and completes the computing power access.

[0099] In the above step S41, the target computing power service is the computing power service that the client needs to access. The address of the computing power service (i.e., CSID) indicates a computing power service, that is, the address of the target computing power service indicates the target computing power service. In an embodiment of the present application, in order to achieve computing power access based on CSID, the CSID is in the same form as the Internet Protocol (IP) address. In one example, the computing power network uses a bearer network based on SRv6 technology as the data plane, and the CSID uses the form of an IPv6 address.

[0100] The types of computing services are limited. To simplify the addresses of computing services and conserve storage resources on devices such as gateways and clients, in an embodiment of the present application, the CSID may consist of a preset address prefix and a computing service identifier (Server Identity, SID). The preset address prefix is ​​a prefix shared by all computing services, meaning that all computing services have the same preset address prefix. The preset address prefix can be set based on actual needs, for example, it can be 0000 or 1111. The SID can indicate a computing service.

[0101] Taking the CSID using an IPv6 address as an example, the structure of the CSID is shown in Figure 5. In Figure 5, the length of the CSID is the same as the length of the IPv6 address, that is, the length of the CSID is 128 bits. The first 12 bytes (96 bits) of the CSID are the preset address prefix, and the last 4 bytes (32 bits) of the CSID are the SID.

[0102] In the embodiment of the present application, the SID can uniquely indicate a computing service, and therefore, the SID can be used to represent the CSID. Since the length of the SID is shorter than that of the CSID, gateway devices and client devices use the SID to represent the CSID, which can greatly save the storage resources occupied when maintaining the CSID.

[0103] When accessing computing power, the client can obtain the address of the target computing power service in the following ways.

[0104] Method 1: The client obtains the address of the target computing power service locally.

[0105] If the client locally records the correspondence between the target computing service name and the target computing service address, the client obtains the target computing service address from the locally recorded correspondence between the target computing service name and the target computing service address based on the target computing service name.

[0106] Method 2: The client obtains the address of the target computing power service through the management layer.

[0107] If the client doesn't have a local record of the mapping between the target computing service's name and address, it can send a domain name resolution request to the management layer's DNS server. The DNS server extracts the target computing service's name from the resolution request and sends it to the management layer's operations and maintenance platform.

[0108] The operation and maintenance platform maintains a computing service information database, which contains the mapping between computing service names and addresses. After obtaining the target computing service name, the operation and maintenance platform queries the computing service information database to obtain the target computing service address and sends this address to the DNS server. The DNS server, in turn, sends the target computing service address to the client. The client receives the target computing service address from the DNS server.

[0109] In the embodiment of the present application, the client may also use other methods to obtain the address of the target computing power service, which is not limited to this.

[0110] In the above step S42, after obtaining the address of the target computing power service, the client uses the client address as the source address and the address of the target computing power service as the destination address to construct a computing power access message, such as the first computing power access message.

[0111] For example, the client's address is IP1, and the target computing service address obtained by the client is CSID1. Based on the obtained target computing service address CSID1, the client generates a computing power access message. The structure of the computing power access message is shown in Figure 6.

[0112] In the above step S43, the target service instance is a service instance that provides the target computing power service, the service node where the target service instance is located is the target service node, the second gateway device can be any gateway device in the computing power network, and the target service node accesses the computing power network through the second gateway device. Specifically, the proxy device of the server where the target service node is located is connected to the second gateway device to enable the target service node to access the computing power network. In order to achieve access to the computing power service provided by the service instance, the address of the service instance (i.e., CSIID) is in the same form as the IP address. In one example, the computing power network uses a bearer network based on SRv6 technology as the data plane, and the CSIID uses the form of an IPv6 address.

[0113] After generating the first computing power access message, the client sends the first computing power access message to the first gateway device. The first gateway device and the second gateway device respectively store a computing power routing table, which is a routing table used to implement computing power access. The computing power routing table includes a CSID, which can be represented by an SID. After receiving the first computing power access message, the first gateway device queries the computing power routing table based on the destination address of the first computing power access message (i.e., the address of the target computing power service), and sends the first computing power access message to the target service node through the second gateway device, so that the client obtains the target computing power service and implements computing power access. The computing power access process based on the computing power routing table will be described in detail later and will not be elaborated here.

[0114] Based on the above computing power access method applied to the client, the embodiment of the present application further provides a computing power access method, as shown in FIG7 , which is applied to the first gateway device. The method includes the following steps:

[0115] Step S71: Receive a first computing power access message sent by a client. The client accesses the computing power network through a first gateway device. The destination address of the first computing power access message is the address of the target computing power service.

[0116] Step S72: Determine the address of the second gateway device corresponding to the address of the target computing service based on a pre-stored first computing power routing table. The first computing power routing table includes the address of the computing service and the address of the gateway device to which the service node where the service instance providing the computing service is located is connected when it accesses the computing power network.

[0117] Step S73: encapsulate the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain a second computing power access message;

[0118] Step S74: Send a second computing power access message to the second gateway device.

[0119] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device in the computing power network stores a computing power routing table containing the address of the computing power service. Based on the address of the target computing power service carried in the computing power access message, the first gateway device queries the computing power routing table and sends the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0120] In the above step S71, the client sends a first computing power access message to the first gateway device, and then the first gateway device receives the first computing power access message. Here, the process of the client sending the first computing power access message to the first gateway device can be referred to the relevant description in the above Figure 4.

[0121] In step S72 above, a computing power routing table, such as the first computing power routing table described above, is pre-stored in the first gateway device. The first computing power routing table may be pre-configured in the first gateway device, or may be generated by the first gateway device based on routing protocol announcements sent by other gateway devices in the computing power network, without limitation. For example, before receiving a first computing power access message sent by a client, the first gateway device receives a routing protocol announcement sent by a second gateway device, the routing protocol announcement including static attribute information and dynamic attribute information of the target service instance; and generates the first computing power routing table based on the static attribute information and the dynamic attribute information.

[0122] In this embodiment of the present application, the first computing power routing table includes the address of the computing power service and the address of the gateway device to which the service node where the service instance providing the computing power service is located connects when accessing the computing power network. The first computing power routing table may also include the address of the service instance providing the computing power service, the routing cost for accessing the service node where the service instance providing the computing power service is located, and static attribute information of the service instance providing the computing power service. The structure of the first computing power routing table is shown in Table 1.

[0123] Table 1

[0124] The SID field is used to fill in the address of the computing service, or the SID indicating the address of the computing service; the CSIID field is used to fill in the address of the service instance; the next hop field is used to fill in the address of the gateway device to which the service node connects when accessing the computing network; the routing cost field is used to fill in the routing cost for accessing the service node, the size of the routing cost can be obtained during the routing protocol announcement; the attribute field is used to fill in computing network information such as static attribute information and dynamic attribute information of the service instance. The computing network information of the service instance can be obtained by the second gateway device from the proxy device of the service node and sent to the first gateway device through the routing protocol announcement.

[0125] After receiving the first computing power access message, the first gateway device extracts the destination address of the first computing power access message, obtains the address of the target computing power service, and then queries the first computing power routing table according to the address of the target computing power service to obtain the address of the gateway device corresponding to the address of the target computing power service. The address of the gateway device is the address of the second gateway device.

[0126] In an embodiment of the present application, in a computing power network, a server connected to a gateway device may contain multiple service instances providing a target computing power service, and multiple gateway devices may be connected to the server providing the target computing power service. In this case, the first gateway device queries the first computing power routing table based on the address of the target computing power service and may obtain multiple routing table entries including the address of the target computing power service.

[0127] At this time, the first gateway device may determine the address of the second gateway device in any of the following ways.

[0128] In mode 1, the first gateway device randomly selects a routing table entry from the obtained multiple routing table entries, and uses the address of the gateway device included in the selected routing table entry as the address of the second gateway device.

[0129] In Method 2, the first computing power routing table also includes the computing cost for accessing the service-providing node. The first gateway device selects the routing entry with the lowest computing cost from the multiple routing table entries obtained and uses the gateway device address included in the selected routing entry as the address of the second gateway device. In other words, the first gateway device determines the address of the second gateway device with the lowest computing cost corresponding to the address of the target computing power service based on the pre-stored first computing power routing table.

[0130] For example, the first computing power routing table in the first gateway device is shown in Table 2.

[0131] Table 2

[0132] The first gateway device obtains the target computing service SID SID1. Based on SID1, it queries Table 2 and obtains three routing entries, namely the first three routing entries in Table 2. The smallest routing cost among these three routing entries is 80. Therefore, the first gateway device selects the third routing entry, which includes the gateway address GW2, which is the address of the second gateway device.

[0133] In an embodiment of the present application, the first gateway device may also determine the address of the second gateway device in other ways, such as determining the address of the second gateway device in combination with the static attribute information included in the first computing power routing table, and this is not limited to this.

[0134] In step S73, the first gateway device encapsulates the first computing power access message based on its own address and the address of the second gateway device determined in step S72. The encapsulated message becomes the second computing power access message. The first gateway device then executes step S74, sending the second computing power access message to the second gateway device via other nodes in the computing power network, thereby accessing the target service node and enabling the client to obtain the target computing power service.

[0135] In some embodiments, the above-mentioned step S73 can be: encapsulating the SRv6 message header in the outer layer of the first computing power access message to obtain a second computing power access message, the source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the forwarding path, and the forwarding path is the path from the first gateway device to the second gateway device.

[0136] Still taking the computing power access message shown in Figure 6 above as an example. The address of the first gateway device is GW1, and the address of the second gateway device is GW2. The first gateway device calculates the forwarding path S from the first gateway device to the second gateway device based on GW1 and GW2, obtains the segment routing list, and then encapsulates the SRv6 message header on the outer layer of the original first computing power access message to obtain the second computing power access message as shown in Figure 8a. In the second computing power access message, the SRv6 message header is the outer header, including the outer IP header and the segment routing header (Segment Routing Header, SRH), and the IP header of the first computing power access message is the inner header. In the computing power network, each routing node on the forwarding path S forwards the second computing power access message based on SRv6 technology until the second gateway device receives the second computing power access message. In Figure 8a, R1 is the address of the next-hop device of the first gateway device on the forwarding path S.

[0137] In some embodiments, the first computing power routing table may further include an identifier of a service instance providing the computing power service, i.e., a CSI ID. In this case, the first gateway device may also determine the address of the target service instance corresponding to the address of the target computing power service based on the pre-stored first computing power routing table.

[0138] Accordingly, in order to accurately control computing power access, the above-mentioned step S73 can be: encapsulating the SRv6 message header in the outer layer of the first computing power access message, and modifying the destination address of the first computing power access message to the address of the target service instance to obtain a second computing power access message, the source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the forwarding path, and the forwarding path is the path from the first gateway device to the second gateway device.

[0139] Still taking the computing power access message shown in Figure 6 above as an example. The address of the first gateway device is GW1, the address of the second gateway device is GW2, and the address of the target service instance is CSIID1. The first gateway device calculates the forwarding path S from the first gateway device to the second gateway device based on GW1 and GW2, obtains the segment routing list, and then encapsulates the SRv6 message header in the outer layer of the original first computing power access message, and modifies the destination address of the first computing power access message to CSIID1, to obtain the second computing power access message shown in Figure 8b. In the second computing power access message, the SRv6 message header is the outer header, including the outer IP header and SRH, and the IP header of the first computing power access message is the inner header. In the computing power network, each routing node on the forwarding path S forwards the second computing power access message based on SRv6 technology until the second gateway device receives the second computing power access message. In Figure 8b, R1 is the address of the next-hop device of the first gateway device on the forwarding path S.

[0140] In some embodiments, in order to more accurately control computing power access and ensure that the client obtains the required computing power service, the second computing power access message may also carry an identifier indicating the address of the target computing power service, such as a CSID or the SID corresponding to the above-mentioned target computing power service.

[0141] In this embodiment of the present application, the SRv6 message header includes an SRH, and the optional field of the SRH is filled with an identifier indicating the address of the target computing service, wherein the optional field of the SRH can be a Type Length Value (TLV) field. As shown in Figure 8c, the optional field of the SRH is filled with SID1 corresponding to CSID1.

[0142] In an embodiment of the present application, the identifier indicating the address of the target computing power service may also be filled in the tail of the second computing power access message, or may be filled in other designated locations in the second computing power access message, without limitation.

[0143] Based on the above computing power access method applied to the client and the first gateway device, the embodiment of the present application also provides a computing power access method, as shown in Figure 9, applied to the second gateway device. The method includes the following steps:

[0144] Step S91: Receive a second computing power access message from a first gateway device. The first gateway device is a gateway device to which the client connects when accessing the computing power network. The second computing power access message is a message obtained by encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device after the first gateway device determines the address of the second gateway device based on a pre-stored first computing power routing table and the destination address of the first computing power access message sent by the client. The destination address of the first computing power access message is the address of the target computing power service. The first computing power routing table includes the address of the computing power service and the address of the gateway device to which the service node where the service instance providing the computing power service is located connects when accessing the computing power network.

[0145] Step S92: decapsulate the second computing power access message to obtain a third computing power access message;

[0146] Step S93: Determine the address of the target proxy device corresponding to the address of the target computing service based on a pre-stored second computing power routing table, where the second computing power routing table includes the address of the computing service and the address of the proxy device that provides the service instance of the computing service.

[0147] Step S94: Send a third computing power access message to the target service node where the target service instance providing the target computing power service is located, according to the address of the target proxy device.

[0148] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0149] In the above step S91, the process of the first gateway device obtaining the second computing power access message can be found in the relevant description of Figures 4 and 7 above, and will not be repeated here.

[0150] After receiving the second computing power access message, the first gateway device sends the second computing power access message to the second gateway device via other nodes in the computing power network. The second gateway device receives the second computing power access message. The other nodes only need to support routing and forwarding and do not need to support the technical solutions provided in the embodiments of the present application.

[0151] In the above step S92, after receiving the second computing power access message, the second gateway device decapsulates the second computing power access message to obtain the inner message of the second computing power access message, that is, the third computing power access message.

[0152] In some embodiments, the first gateway device encapsulates an SRv6 header on the outer layer of the first computing power access message to obtain a second computing power access message. The source address of the SRv6 header is the address of the first gateway device, and the destination address of the SRv6 header is the address of the next-hop device of the first gateway device on the path from the first gateway device to the second gateway device. In this case, based on SRv6 technology, the second computing power access message is forwarded in the computing power network, and the destination address of the second computing power access message remains unchanged. When the second gateway device receives the second computing power access message, the second computing power access message includes an SRv6 header and a payload. The source address of the SRv6 header is the address of the first gateway device, the destination address of the SRv6 header is the address of the second gateway device, and the payload is the first computing power access message.

[0153] At this time, the above step S92 may be: stripping the SRv6 message header of the second computing power access message to obtain the first computing power access message, where the first computing power access message is the third computing power access message.

[0154] In some embodiments, the first computing power routing table may further include the address of a service instance providing computing power services; the second computing power routing table may further include the address of a service instance providing computing power services. The first gateway device encapsulates an SRv6 header on the outer layer of the first computing power access message, and modifies the destination address of the first computing power access message to the address of the target service instance, thereby obtaining a second computing power access message, wherein the source address of the SRv6 header is the address of the first gateway device, and the destination address of the SRv6 header is the address of the next hop device of the first gateway device on the path from the first gateway device to the second gateway device.

[0155] In this case, based on SRv6 technology, the second computing power access message is forwarded in the computing power network, and the destination address of the second computing power access message remains unchanged. When the second gateway device receives the second computing power access message, the second computing power access message includes an SRv6 message header and a payload. The source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is the computing power access message obtained by modifying the destination address of the first computing power access message to the address of the target service instance, that is, the third computing power access message;

[0156] At this time, the above step S92 may be: stripping the SRv6 message header of the second computing power access message to obtain the third computing power access message.

[0157] In the above step S93, the target proxy device is a proxy device of the computing resource pool where the target service instance is located.

[0158] The second gateway device pre-stores a computing power routing table, such as the second computing power routing table described above. The second computing power routing table can be pre-configured in the second gateway device.

[0159] To conserve network resources during computing power access, the second computing power routing table can also be generated by the second gateway device based on computing power resource pool information collected by the target proxy device. For example, before receiving the second computing power access message, the second gateway device receives static attribute information and dynamic attribute information of the target service instance collected by the target proxy device; and generates the second computing power routing table based on the static attribute information and dynamic attribute information.

[0160] To conserve table entry resources in the second gateway device, the second computing power routing table can also be issued by the operation and maintenance platform when the client accesses computing power. For example, before receiving the second computing power access message, the second gateway device receives the second computing power routing table issued by the operation and maintenance platform of the computing power network and stores the second computing power routing table. The second computing power routing table in the operation and maintenance platform is generated based on the static and dynamic attribute information of the target service instance reported by the second gateway device.

[0161] The second computing power routing table may also be obtained through other methods, which are not limited thereto.

[0162] In this embodiment of the present application, the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service. The second computing power routing table may also include the address of the service instance providing the computing power service, the routing cost for accessing the service node where the service instance providing the computing power service is located, and static attribute information of the service instance providing the computing power service. The structure of the second computing power routing table is shown in Table 3.

[0163] Table 3

[0164] The SID field is used to fill in the address of the computing service, or the SID indicating the address of the computing service; the CSIID field is used to fill in the address of the service instance; the next hop field is used to fill in the address of the proxy device; the routing cost field is used to fill in the routing cost of accessing the service node, the size of the routing cost can be obtained during the routing protocol announcement; the attribute field is used to fill in computing network information such as static and dynamic attribute information of the service instance. The computing network information of the service instance can be obtained by the second gateway device from the proxy device of the service node.

[0165] In some embodiments, when the second computing power access message carries the address of the target service instance, if the destination address of the third computing power access message is the address of the target service instance, the above step S93 can be to determine the address of the target proxy device corresponding to the address of the target service instance based on the pre-stored second computing power routing table.

[0166] In an embodiment of the present application, multiple service instances can provide the same computing power service. A service instance provides a computing power service. That is, the target service instance provides the target computing power service, and the address of the target proxy device corresponding to the address of the target service instance is the address of the target proxy device corresponding to the address of the target computing power service. Based on the address of the target service instance, the second gateway device can more accurately determine the address of the target proxy device, thereby improving the accuracy of computing power access.

[0167] In some embodiments, to further improve the accuracy of computing power access, the second computing power access message may also carry a target identifier representing the address of the target computing power service, such as the above-mentioned SID. In an embodiment of the present application, the SRv6 message header includes an SRH, and the optional field of the SRH is filled with the target identifier. The target identifier may also be filled in the tail of the second computing power access message, or in other specified locations in the second computing power access message, without limitation.

[0168] In this case, the second gateway device can also extract the target identifier from the second computing power access message and, based on a pre-stored second computing power routing table, determine the address of the target proxy device corresponding to the target identifier and the address of the target service instance. Combining multiple pieces of information to determine the target proxy device address further improves the accuracy of computing power access.

[0169] In some embodiments, a method for allocating addresses for a computing power service is also provided, as shown in FIG10 , which may include the following steps.

[0170] Step S101: Receive static attribute information of a target service instance collected by a target proxy device, where the static attribute information includes the name of the target computing service.

[0171] In an embodiment of the present application, the target proxy device may periodically collect static attribute information of the target service instance and, upon detecting a change in the static attribute information of the target service instance, send the static attribute information to the second gateway device. The target proxy device may also collect static attribute information of the target service instance and send the static attribute information to the second gateway device upon receiving a notification event related to information reporting. The target proxy device may also use other methods to collect static attribute information of the target service instance, which are not limited to this method.

[0172] Step S102: If the second gateway device does not store the address of the target computing service corresponding to the name of the target computing service, static attribute information is sent to the operation and maintenance platform of the computing network.

[0173] In an embodiment of the present application, the second gateway device maintains a correspondence between the name of the computing service and the address of the computing service. After obtaining the static attribute information of the target service instance, the second gateway device determines whether the second gateway device stores the address of the target computing service based on the name of the target computing service included in the static attribute information, in combination with the maintained correspondence.

[0174] If the address of the target computing power service is stored, it means that the operation and maintenance platform has allocated an address for the target computing power service. The second gateway device ends the address allocation process and performs subsequent operations, such as sending a routing protocol announcement to the first gateway device. The routing protocol announcement includes static attribute information and dynamic attribute information of the target service instance, so that the first gateway device can generate a first computing power routing table.

[0175] If the address of the target computing service is not stored, it means that the operation and maintenance platform has not yet assigned an address to the target computing service. At this time, the second gateway device can send static attribute information of the target service instance to the operation and maintenance platform.

[0176] Step S103: Receive the address of the target computing power service allocated by the operation and maintenance platform to the target computing power service according to the static attribute information.

[0177] Based on the static attribute information of the target service instance, the operation and maintenance platform can obtain the target computing service's name, service type, and other information, and then assign an address to the target computing service, namely the target computing service's CSID. The operation and maintenance platform feeds the target computing service's address back to the second gateway device, which records and maintains the correspondence between the target computing service's name and address.

[0178] In addition, after obtaining the address of the target computing power service, the second gateway device can perform subsequent operations, such as sending a routing protocol announcement to the first gateway device. The routing protocol announcement includes static attribute information and dynamic attribute information of the target service instance, so that the first gateway device can generate a first computing power routing table.

[0179] Through the embodiment shown in Figure 10, the address of the target computing service is allocated, ensuring that subsequent computing power access based on the address of the target computing service is achieved. In addition, the second gateway device maintains the correspondence between the computing service name and the computing service address. For computing services that have already been assigned addresses, the second gateway device no longer requests the operation and maintenance platform to allocate an address for the computing service, thus saving network resources and the computing resources of the operation and maintenance platform.

[0180] In addition, the operation and maintenance platform assigns addresses to computing power services and collects static attribute information of service instances to facilitate the generation of the first computing power routing table and the second computing power routing table, and then configures the first gateway device and the second gateway device to facilitate computing power access.

[0181] Based on the above-mentioned computing power access method applied to the client, the first gateway device, and the second gateway device, the embodiment of the present application also provides a computing power access method, as shown in Figure 11, which is applied to the operation and maintenance platform. The method includes the following steps:

[0182] Step S111, obtaining the name of the target computing service from the client;

[0183] Step S112: determining the address of the target computing service corresponding to the name of the target computing service based on the pre-stored correspondence between the names of the computing services and the addresses of the computing services;

[0184] Step S113: Send the address of the target computing service to the client, so that the client can access the target service node where the target service instance providing the target computing service is located through the first gateway device and the second gateway device in the computing network according to the address of the target computing service. The client accesses the computing network through the first gateway device, and the target service node accesses the computing network through the second gateway device.

[0185] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0186] In step S111, the client can directly connect to the operation and maintenance platform. When accessing computing power, the client sends the name of the target computing power service to the operation and maintenance platform.

[0187] Clients can also connect to the operation and maintenance platform through a DNS server. When accessing computing power, the client can send a domain name resolution request to the DNS server. The domain name included in the domain name resolution request is the name of the target computing power service. The DNS server extracts the name of the target computing power service from the domain name resolution request and sends the extracted target computing power service name to the operation and maintenance platform.

[0188] In step S112, the operation and maintenance platform maintains a correspondence between the name of the computing service and the address of the computing service. After obtaining the name of the target computing service, the operation and maintenance platform can query the correspondence between the name of the computing service and the address of the computing service to obtain the address of the computing service corresponding to the name of the target computing service, that is, the address of the target computing service.

[0189] In an embodiment of the present application, the correspondence between the name of the computing power service maintained by the operation and maintenance platform and the address of the computing power service can be determined based on the static attribute information reported by the second gateway device. For example, before obtaining the name of the target computing power service from the client, the operation and maintenance platform obtains the static attribute information of the target service instance from the second gateway device, and the static attribute information includes the name of the target computing power service. According to the static attribute information, the target computing power service is assigned the address of the target computing power service; and the address of the target computing power service is sent to the second gateway device. At this time, the operation and maintenance platform obtains the correspondence between the name of the target computing power service and the address of the target computing power service. Subsequently, when the operation and maintenance platform again obtains the static attribute information including the name of the target computing power service reported by other gateway devices, it can directly send the address of the target computing power service to other gateway devices based on the above correspondence. Here, the address allocation process of the computing power service can refer to the relevant description of the above Figure 10.

[0190] In step S113, the operation and maintenance platform sends the obtained address of the target computing service to the client. The client then accesses the target service instance through the first gateway device and the second gateway device. For the specific access process, please refer to the relevant descriptions of Figures 4, 7, and 9 above.

[0191] In some embodiments, to ensure the realization of computing power access, the operation and maintenance platform can maintain the correspondence between the address of the computing power service and the address of the service instance providing the computing power service based on the computing network information reported by the second gateway device. After determining the address of the target computing power service, the operation and maintenance platform can determine the address of the target service instance corresponding to the address of the target computing power service based on the pre-stored correspondence between the address of the computing power service and the address of the service instance; and then generate a second computing power routing table from the second gateway device to the target service instance based on the address of the target service instance, the second computing power routing table including the address of the computing power service and the address of the proxy device of the service instance providing the computing power service; and send the second computing power routing table to the second gateway device.

[0192] In this way, the second gateway device does not need to generate the second computing power routing table in advance, and generates the second computing power routing table when it needs to access the target service instance, which saves the table entry resources of the second gateway device and also ensures the realization of computing power access.

[0193] The computing power access method provided in the embodiment of the present application is described in detail below in conjunction with the computing power access system shown in FIG3 and the computing power access signaling diagrams shown in FIG12 and FIG13 .

[0194] In the preparation stage of computing power access, as shown in Figure 12.

[0195] In step S121 , the proxy device 324 collects static attribute information of each service instance in the server 325 .

[0196] The static attribute information may include the address of the computing service within the server 325, the name of the computing service, the service type of the computing service, the address of the service instance, the location of the service instance, the priority of the service instance, and the capabilities of the service instance. The capabilities of the service instance may include codec capabilities, image processing capabilities, etc. The address of the service instance (i.e., CSIID) included in the static attribute information is the IP address of the service instance.

[0197] In step S122 , the proxy device 324 notifies the gateway device 323 of the collected static attribute information.

[0198] In step S123 , the gateway device 323 notifies the network controller 313 of the received static attribute information.

[0199] In step S124 , the network controller 313 notifies the received static attribute information to the operation and maintenance platform 311 .

[0200] In step S125, the operation and maintenance platform 311 allocates a CSID to the computing power service based on the received static attribute information.

[0201] In step S126 , the operation and maintenance platform 311 returns the allocated CSID to the network controller 313 .

[0202] In step S127 , the network controller 313 returns the received CSID to the gateway device 323 .

[0203] In step S128, the gateway device 323 maintains the correspondence between the name of the computing power service and the CSID based on the received CSID.

[0204] In step S129, the proxy device 324 collects dynamic attribute information of each service instance in the server 325. The dynamic attribute information may include load status and session status.

[0205] In step S1210 , the proxy device 324 notifies the gateway device 323 of the collected dynamic attribute information.

[0206] In step S1211, the gateway device 323 notifies the received static attribute information and dynamic attribute information to other gateway devices such as the gateway device 322 through a routing protocol. The routing protocol may be the Border Gateway Protocol (BGP) or the Address Resolution Protocol (ARP).

[0207] In step S1212, the gateway device 322 maintains a local computing power routing table based on the collected static attribute information and dynamic attribute information.

[0208] Since each gateway device may be an egress gateway device (such as the second gateway device) or an ingress gateway device (such as the first gateway device), the local computing power routing table of each gateway device may include the information in the above-mentioned first computing power routing table and the information in the second computing power routing table.

[0209] During the execution phase of computing power access, as shown in Figure 13.

[0210] In step S131, the client 321 sends a domain name resolution request to the DNS server 312. The domain name carried in the domain name resolution request is the name of the target computing power service.

[0211] In step S132, the DNS server 312 requests service IP resolution from the operation and maintenance platform 311. Here, the service IP is the address CSID of the target computing power service.

[0212] The DNS server 312 requests service IP resolution from the operation and maintenance platform 311 , that is, sends the name of the target computing power service to the operation and maintenance platform 311 .

[0213] In step S133, the operation and maintenance platform 311 queries the registered computing power service information database to obtain the service instance that provides the target computing power service. Here, the operation and maintenance platform 311 can also obtain the service IP, that is, the address CSID of the target computing power service.

[0214] In step S134, the operation and maintenance platform 311 sends the policy corresponding to the service instance of the target computing power service to the network controller 313. The policy includes the second computing power routing table. After obtaining the service instance that provides the target computing power service, the operation and maintenance platform 311 can obtain the policy corresponding to the service instance.

[0215] In step S135 , the network controller 313 sends the received policy to the gateway device 323 where the service instance providing the target computing power service is located.

[0216] In step S136, the operation and maintenance platform 311 returns the service IP to the DNS server 312. The embodiment of the present application does not limit the execution order of step S134 and step S136.

[0217] In step S137 , the DNS server 312 returns the service IP to the client 321 .

[0218] In step S138, the client 321 sends a computing power access message to the gateway device 322. The computing power access message is an IPv6 message, and the destination IP is the service IP.

[0219] In step S139, after receiving the computing power access message, gateway device 322 extracts the CSID (i.e., service IP address) from the destination IP address of the computing power access message, queries the computing power routing table based on the CSID, selects the optimal routing table entry, and calculates the path from gateway device 322 to gateway device 323. Here, gateway device 323 is the gateway device where the service instance indicated by the routing table entry is located.

[0220] For example, gateway device 322 extracts the SID from the CSID and encapsulates the SID in the optional TLV field of the SRH. Based on the extracted SID, gateway device 322 queries the computing power routing table and obtains the routing entry with the lowest routing cost. This routing entry is the optimal routing entry, which includes the CSIID and the next hop address (i.e., the address of gateway device 323). Based on the next hop address and the address of gateway device 322 itself, gateway device 322 can calculate an optimal SRv6 path from gateway device 322 to gateway device 323.

[0221] In step S1310, the gateway device 322 adds SRv6 encapsulation to the outer layer of the computing power access message, modifies the destination IP of the computing power access message to CSIID, and encapsulates CSID in the optional TLV field of SRH to obtain a new computing power access message.

[0222] For example, the gateway device 322 extracts the SID from the CSID and encapsulates the SID in the optional TLV field of the SRH.

[0223] In step S1311 , the gateway device 322 sends a new computing power access message to the gateway device 323 .

[0224] In step S1312, after receiving the computing power access message, the gateway device 323 decapsulates the outer layer of the computing power access message and extracts the CSID from the optional TLV field of the SRH.

[0225] The gateway device 323 decapsulates the outer SRv6 message header (including the IP header and SRH) of the computing power access message to obtain the SID and the inner message.

[0226] In step S1313, the gateway device 323 queries the computing power routing table and forwards the inner layer message to the service node in the server 325 where the service instance is located.

[0227] Gateway device 323 queries the computing power routing table based on the SID obtained in step S1311 and the destination address CSIID of the inner message, and obtains the next hop address, which is the address of proxy device 324. Based on the address of proxy device 324, gateway device 323 then sends the inner message to the service node where the service instance in server 325 is located, completing the computing power access.

[0228] The description of the above Figures 12 and 13 is relatively simple. For details, please refer to the relevant description of the above Figures 4 to 11.

[0229] The technical solution provided in the embodiment of this application adopts a hybrid forwarding mechanism, that is, it cooperates with the upper-level control layer and the lower-level device layer to complete computing power access. In addition, the technical solution provided in the embodiment of this application does not require modifying the DNS mechanism, and can directly use the DNS mechanism to complete computing power access, reducing the deployment cost of the computing power network.

[0230] Corresponding to the above-mentioned computing power access method, an embodiment of the present application further provides a computing power access device, as shown in FIG14 , which is applied to a first gateway device, and includes:

[0231] The first receiving module 141 is configured to receive a first computing power access message sent by a client, wherein the client accesses the computing power network through a first gateway device, and the destination address of the first computing power access message is the address of a target computing power service;

[0232] Determining module 142, configured to determine the address of a second gateway device corresponding to the address of the target computing service based on a pre-stored first computing power routing table, where the first computing power routing table includes the address of the computing service and the address of the gateway device to which the service node where the service instance providing the computing service is located is connected when the service node accesses the computing power network;

[0233] An encapsulation module 143 is configured to encapsulate the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain a second computing power access message;

[0234] The sending module 144 is configured to send a second computing power access message to the second gateway device.

[0235] In some embodiments, the first computing power routing table further includes a routing cost for accessing a service node where a service instance providing computing power service is located;

[0236] The determination module 142 may be specifically configured to determine the address of the second gateway device with the minimum routing cost corresponding to the address of the target computing power service according to a pre-stored first computing power routing table.

[0237] In some embodiments, the encapsulation module 143 can be specifically used to encapsulate the SRv6 message header in the outer layer of the first computing power access message to obtain a second computing power access message, the source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the path from the first gateway device to the second gateway device.

[0238] In some embodiments, the first computing power routing table further includes an address of a service instance providing computing power services;

[0239] The determination module 142 may also be configured to determine the address of the target service instance corresponding to the address of the target computing service based on the pre-stored first computing power routing table;

[0240] The encapsulation module 143 can be specifically used to encapsulate the SRv6 message header in the outer layer of the first computing power access message, and modify the destination address of the first computing power access message to the address of the target service instance to obtain a second computing power access message. The source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the path from the first gateway device to the second gateway device.

[0241] In some embodiments, the SRv6 message header includes an SRH, and an optional field of the SRH is filled with an identifier indicating the address of the target computing power service.

[0242] In some embodiments, the computing power access device may further include:

[0243] A second receiving module is configured to receive a routing protocol announcement sent by a second gateway device before receiving the first computing power access message sent by the client, where the routing protocol announcement includes static attribute information and dynamic attribute information of a target service instance that provides a target computing power service;

[0244] A generation module is used to generate a first computing power routing table based on static attribute information and dynamic attribute information.

[0245] In some embodiments, the static attribute information may include one or more of the address of the target computing service, the name of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance, and the capability of the target service instance; or

[0246] The dynamic attribute information may include one or more of a load state and a session state of the target service instance.

[0247] In some embodiments, the address of the computing power service consists of a preset address prefix and an identifier of the computing power service;

[0248] The address of the computing power service included in the first computing power routing table is represented by the identifier of the computing power service.

[0249] In some embodiments, the address of the computing service and the address of the service instance are respectively in the same form as IP addresses.

[0250] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0251] Corresponding to the above-mentioned computing power access method, an embodiment of the present application further provides a computing power access device, as shown in FIG15 , which is applied to a second gateway device, and includes:

[0252] The first receiving module 151 is used to receive a second computing power access message from the first gateway device, where the first gateway device is the gateway device to which the client connects when accessing the computing power network. The second computing power access message is a message obtained by encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device after the first gateway device determines the address of the second gateway device based on a pre-stored first computing power routing table and the destination address of the first computing power access message sent by the client. The destination address of the first computing power access message is the address of the target computing power service. The first computing power routing table includes the address of the computing power service and the address of the gateway device to which the service node where the service instance providing the computing power service is located connects when accessing the computing power network.

[0253] a decapsulation module 152, configured to decapsulate the second computing power access message to obtain a third computing power access message;

[0254] A determination module 153 is configured to determine an address of a target proxy device corresponding to an address of a target computing service based on a pre-stored second computing power routing table, wherein the second computing power routing table includes the address of the computing service and the address of a proxy device that provides a service instance of the computing service;

[0255] The first sending module 154 is configured to send a third computing power access message to a service node where a target service instance providing a target computing power service is located according to an address of a target proxy device.

[0256] In some embodiments, the second computing power access message includes an SRv6 message header and a payload, the source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is the first computing power access message;

[0257] The decapsulation module 152 may be specifically configured to strip the SRv6 message header of the second computing power access message to obtain the first computing power access message, where the first computing power access message is the third computing power access message.

[0258] In some embodiments, the first computing power routing table may further include an address of a service instance providing computing power services;

[0259] The second computing power routing table may further include an address of a service instance providing computing power services;

[0260] The second computing power access message may include an SRv6 message header and a payload. The source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is a third computing power access message obtained by modifying the destination address of the first computing power access message to the address of the target service instance.

[0261] The decapsulation module 152 may be specifically configured to strip the SRv6 header of the second computing power access message to obtain a third computing power access message;

[0262] The determination module 153 may be specifically configured to determine the address of the target proxy device corresponding to the address of the target service instance according to a pre-stored second computing power routing table.

[0263] In some embodiments, the SRv6 message header includes an SRH, and an optional field of the SRH is filled with a target identifier indicating the address of the target computing power service;

[0264] The determination module is specifically used to determine the address of the target proxy device corresponding to the target identifier and the address of the target service instance according to the pre-stored second computing power routing table.

[0265] In some embodiments, the computing power access device may further include:

[0266] A second receiving module is configured to receive a second computing power routing table issued by an operation and maintenance platform in the computing power network before receiving the second computing power access message;

[0267] The storage module is used to store the second computing power routing table.

[0268] In some embodiments, the computing power access device may further include:

[0269] A third receiving module is configured to receive static attribute information and dynamic attribute information of a target service instance collected by a target proxy device before receiving the second computing power access message;

[0270] A generation module is used to generate a second computing power routing table based on the static attribute information and the dynamic attribute information.

[0271] In some embodiments, the dynamic attribute information may include one or more of a load status and a session status of the target service instance.

[0272] In some embodiments, the computing power access device may further include:

[0273] a fourth receiving module, configured to receive static attribute information of a target service instance collected by a target proxy device before receiving the second computing power access message, the static attribute information including a name of the target computing power service;

[0274] A second sending module is configured to send static attribute information to an operation and maintenance platform in the computing power network if the second gateway device does not store the address of the target computing power service corresponding to the name of the target computing power service;

[0275] The fifth receiving module is used to receive the address of the target computing power service allocated by the operation and maintenance platform to the target computing power service according to the static attribute information.

[0276] In some embodiments, the computing power access device may further include:

[0277] The third sending module is used to send a routing protocol announcement to the first gateway device after obtaining the address of the target computing power service. The routing protocol announcement includes static attribute information and dynamic attribute information of the target service instance.

[0278] In some embodiments, the static attribute information may include one or more of the address of the target computing service, the name of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance, and the capabilities of the target service instance.

[0279] In some embodiments, the address of the computing power service consists of a preset address prefix and an identifier of the computing power service;

[0280] The addresses of the computing power services included in the first computing power routing table and the second computing power routing table are both represented by the identifiers of the computing power services.

[0281] In some embodiments, the address of the computing service and the address of the service instance are respectively in the same form as IP addresses.

[0282] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0283] Corresponding to the above-mentioned computing power access method, an embodiment of the present application further provides a computing power access device, as shown in FIG16 , which is applied to an operation and maintenance platform, and includes:

[0284] The first acquisition module 161 is used to obtain the name of the target computing service from the client;

[0285] A first determining module 162 is configured to determine the address of a target computing service corresponding to the name of the target computing service based on a pre-stored correspondence between the names of computing services and the addresses of computing services;

[0286] The first sending module 163 is used to send the address of the target computing power service to the client, so that the client can access the target service node where the target service instance providing the target computing power service is located through the first gateway device and the second gateway device of the computing power network according to the address of the target computing power service. The client accesses the computing power network through the first gateway device, and the target service node accesses the computing power network through the second gateway device.

[0287] In some embodiments, the first acquisition module 161 can be specifically used to receive the name of the target computing service sent by the DNS server. The name of the target computing service is the domain name extracted by the DNS server from the domain name resolution request sent by the client.

[0288] In some embodiments, the computing power access device may further include:

[0289] A second determining module is configured to, after determining the address of the target computing service, determine the address of the target service instance corresponding to the address of the target computing service based on a pre-stored correspondence between the address of the computing service and the address of the service instance providing the computing service;

[0290] A generation module, configured to generate a second computing power routing table from the second gateway device to the target service node according to the address of the target service instance, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service;

[0291] The second sending module is used to send the second computing power routing table to the second gateway device.

[0292] In some embodiments, the computing power access device may further include:

[0293] A second acquisition module is configured to acquire static attribute information of the target service instance from the second gateway device before acquiring the name of the target computing service from the client, where the static attribute information includes the name of the target computing service;

[0294] An allocation module, configured to allocate the address of a target computing service to the target computing service based on static attribute information;

[0295] The third sending module is used to send the address of the target computing power service to the second gateway device.

[0296] In some embodiments, the static attribute information may also include one or more of the address of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance, and the capabilities of the target service instance.

[0297] In some embodiments, the address of the computing service and the address of the service instance are respectively in the same form as IP addresses.

[0298] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0299] Corresponding to the above-mentioned computing power access method, an embodiment of the present application further provides a computing power access device, as shown in FIG17 , which is applied to a client, and the client accesses the computing power network through a first gateway device. The device includes:

[0300] Acquisition module 171, used to obtain the address of the target computing power service;

[0301] A generating module 172 is configured to generate a first computing power access message whose destination address is an address of a target computing power service;

[0302] The sending module 173 is used to send a first computing power access message to the first gateway device. The client accesses the computing power network through the first gateway device, so that the first gateway device sends the first computing power access message to the target service node where the target service instance providing the target computing power service is located through the second gateway device. The target service node accesses the computing power network through the second gateway device.

[0303] In some embodiments, the acquisition module 171 may be specifically configured to:

[0304] Send a domain name resolution request to the DNS server. The domain name included in the domain name resolution request is the name of the target computing power service.

[0305] The target computing service address corresponding to the target computing service name received from the DNS server.

[0306] In some embodiments, the address of the target computing service consists of a preset address prefix and an identifier of the target computing service;

[0307] In the computing power routing tables stored in the first gateway device and the second gateway device, the address of the target computing power service is represented by the identifier of the target computing power service.

[0308] In some embodiments, the address of the target computing service and the address of the target service instance are respectively in the same form as IP addresses.

[0309] In the technical solution provided by the embodiment of the present application, an address is assigned to a computing power service. The client constructs a computing power access message with the destination address being the address of the target computing power service, and sends the computing power access message to the first gateway device of the computing power network. The first gateway device and the second gateway device in the computing power network respectively store a computing power routing table containing the address of the computing power service. The first gateway device and the second gateway device query the computing power routing table based on the address of the target computing power service carried in the computing power access message, and send the computing power access message to the target service node where the target service instance providing the target computing power service is located, thereby realizing data plane forwarding and completing computing power access.

[0310] Corresponding to the above-mentioned computing power access method, an embodiment of the present application also provides a network device, as shown in Figure 18, including a processor 181 and a machine-readable storage medium 182, wherein the machine-readable storage medium 182 stores machine-executable instructions that can be executed by the processor 181, and the processor 181 is prompted by the machine-executable instructions to: implement any of the above-mentioned computing power access methods applied to the first gateway device, or implement any of the above-mentioned computing power access methods applied to the second gateway device, or implement any of the above-mentioned computing power access methods applied to the operation and maintenance platform, or implement any of the above-mentioned computing power access methods applied to the client.

[0311] The machine-readable storage medium may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the machine-readable storage medium may be at least one storage device located remote from the processor.

[0312] The processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0313] In another embodiment provided in the present application, a computer-readable storage medium is also provided, which stores a computer program. When the computer program is executed by the processor, it implements any of the above-mentioned computing power access methods applied to the first gateway device, or implements any of the above-mentioned computing power access methods applied to the second gateway device, or implements any of the above-mentioned computing power access methods applied to the operation and maintenance platform, or implements any of the above-mentioned computing power access methods applied to the client.

[0314] In another embodiment provided in the present application, a computer program product comprising instructions is also provided. When the computer is run on a computer, the computer executes any of the computing power access methods applied to the first gateway device, or executes any of the computing power access methods applied to the second gateway device, or executes any of the computing power access methods applied to the operation and maintenance platform, or executes any of the computing power access methods applied to the client.

[0315] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0316] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0317] Each embodiment in this specification is described in a related manner. Similar portions between embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the apparatus, network device, storage medium, and program product embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For related portions, reference can be made to the descriptions of the method embodiments.

[0318] The above description is only a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application are included in the scope of protection of the present application.

Claims

1. A computing power access method, characterized in that: Applied to a first gateway device, the method includes: Receive a first computing power access message sent by a client, the client accessing the computing power network through the first gateway device, and the destination address of the first computing power access message is the address of the target computing power service; Determining, based on a pre-stored first computing power routing table, an address of a second gateway device corresponding to the address of the target computing power service, wherein the first computing power routing table includes the address of the computing power service and the address of a gateway device to which a service node where a service instance providing the computing power service is located is connected when the service node accesses the computing power network; Encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain a second computing power access message; Send the second computing power access message to the second gateway device.

2. The method according to claim 1, characterized in that The first computing power routing table also includes a routing cost for accessing a service node where a service instance providing computing power services is located; The step of determining the address of the second gateway device corresponding to the address of the target computing power service according to the pre-stored first computing power routing table includes: According to the pre-stored first computing power routing table, the address of the second gateway device with the minimum routing cost corresponding to the address of the target computing power service is determined.

3. The method according to claim 1, characterized in that The step of encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain the second computing power access message includes: A segment routing SRv6 message header based on the sixth version of the Internet protocol is encapsulated in the outer layer of the first computing power access message to obtain a second computing power access message, where the source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the path from the first gateway device to the second gateway device.

4. The method according to claim 1, wherein The first computing power routing table also includes an address of a service instance providing computing power services; the method further includes: Determine, according to a pre-stored first computing power routing table, an address of a target service instance corresponding to the address of the target computing power service; The step of encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain the second computing power access message includes: An SRv6 message header is encapsulated in the outer layer of the first computing power access message, and the destination address of the first computing power access message is modified to the address of the target service instance to obtain a second computing power access message, where the source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the path from the first gateway device to the second gateway device.

5. The method according to claim 4, characterized in that The SRv6 message header includes a segment routing header (SRH), and an optional field of the SRH is filled with an identifier indicating the address of the target computing power service.

6. The method according to claim 1, characterized in that Before receiving the first computing power access message sent by the client, the method further includes: Receive a routing protocol announcement sent by the second gateway device, where the routing protocol announcement includes static attribute information and dynamic attribute information of a target service instance that provides the target computing power service; Generate the first computing power routing table according to the static attribute information and the dynamic attribute information.

7. The method according to claim 6, characterized in that The static attribute information includes one or more of the address of the target computing service, the name of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance, and the capability of the target service instance; or The dynamic attribute information includes one or more of a load state and a session state of the target service instance.

8. The method according to any one of claims 1 to 7, characterized in that The address of the computing power service consists of a preset address prefix and an identifier of the computing power service; The address of the computing power service included in the first computing power routing table is represented by an identifier of the computing power service.

9. The method according to any one of claims 1 to 7, characterized in that The address of the computing power service and the address of the service instance are respectively in the same format as the network protocol IP address.

10. A computing power access method, characterized in that: Applied to a second gateway device, the method includes: Receive a second computing power access message from a first gateway device, where the first gateway device is the gateway device connected when the client accesses the computing power network. The second computing power access message is a message obtained by encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device after the first gateway device determines the address of the second gateway device based on a pre-stored first computing power routing table and the destination address of the first computing power access message sent by the client. The destination address of the first computing power access message is the address of the target computing power service. The first computing power routing table includes the address of the computing power service and the address of the gateway device connected when the service node where the service instance providing the computing power service is located accesses the computing power network. Decapsulating the second computing power access message to obtain a third computing power access message; Determining, based on a pre-stored second computing power routing table, an address of a target proxy device corresponding to the address of the target computing power service, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service; The third computing power access message is sent, according to the address of the target proxy device, to the target service node where the target service instance providing the target computing power service is located.

11. The method according to claim 10, characterized in that The second computing power access message includes a segment routing (SRv6) message header and a payload based on Internet Protocol version 6, where the source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is the first computing power access message; The step of decapsulating the second computing power access message to obtain a third computing power access message includes: The SRv6 message header of the second computing power access message is stripped to obtain the first computing power access message, where the first computing power access message is the third computing power access message.

12. The method according to claim 10, characterized in that The first computing power routing table further includes an address of a service instance providing computing power services; the second computing power routing table further includes an address of a service instance providing computing power services; The second computing power access message includes an SRv6 message header and a payload, the source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is a third computing power access message obtained by modifying the destination address of the first computing power access message to the address of the target service instance; The step of decapsulating the second computing power access message to obtain a third computing power access message includes: Stripping the SRv6 message header from the second computing power access message to obtain a third computing power access message; The step of determining the address of the target proxy device corresponding to the address of the target computing power service according to the pre-stored second computing power routing table includes: According to the pre-stored second computing power routing table, the address of the target proxy device corresponding to the address of the target service instance is determined.

13. The method according to claim 12, characterized in that The SRv6 message header includes a segment routing header (SRH), and an optional field of the SRH is filled with a target identifier indicating the address of the target computing power service; The step of determining the address of the target proxy device corresponding to the address of the target service instance according to the pre-stored second computing power routing table includes: According to the pre-stored second computing power routing table, the address of the target proxy device corresponding to the target identifier and the address of the target service instance is determined.

14. The method according to claim 10, characterized in that Before receiving the second computing power access message, the method further includes: Receiving the second computing power routing table issued by the operation and maintenance platform in the computing power network; The second computing power routing table is stored.

15. The method according to claim 10, characterized in that Before receiving the second computing power access message, the method further includes: receiving static attribute information and dynamic attribute information of the target service instance collected by the target proxy device; Generate the second computing power routing table according to the static attribute information and the dynamic attribute information.

16. The method according to claim 15, characterized in that The dynamic attribute information includes one or more of a load state and a session state of the target service instance.

17. The method according to claim 10, wherein: Before receiving the second computing power access message, the method further includes: Receiving static attribute information of the target service instance collected by the target proxy device, the static attribute information including the name of the target computing service; If the second gateway device does not store the address of the target computing service corresponding to the name of the target computing service, sending the static attribute information to the operation and maintenance platform in the computing network; Receive the address of the target computing power service allocated by the operation and maintenance platform to the target computing power service according to the static attribute information.

18. The method according to claim 17, characterized in that After obtaining the address of the target computing power service, the method further includes: A routing protocol announcement is sent to the first gateway device, where the routing protocol announcement includes static attribute information and dynamic attribute information of the target service instance.

19. The method according to any one of claims 15 to 18, characterized in that: The static attribute information includes one or more of the address of the target computing service, the name of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance and the capabilities of the target service instance.

20. The method according to any one of claims 10 to 18, characterized in that: The address of the computing power service consists of a preset address prefix and an identifier of the computing power service; The addresses of the computing power services included in the first computing power routing table and the second computing power routing table are both represented by identifiers of the computing power services.

21. The method according to any one of claims 10 to 18, wherein: The address of the computing power service and the address of the service instance are respectively in the same format as the network protocol IP address.

22. A computing power access method, characterized in that: Applied to the operation and maintenance platform, the method includes: Get the name of the target computing service from the client; Determine the address of the target computing service corresponding to the name of the target computing service according to the pre-stored correspondence between the name of the computing service and the address of the computing service; The address of the target computing service is sent to the client, so that the client can access the target service node where the target service instance providing the target computing service is located through the first gateway device and the second gateway device in the computing network according to the address of the target computing service. The client accesses the computing network through the first gateway device, and the target service node accesses the computing network through the second gateway device.

23. The method according to claim 22, characterized in that The step of obtaining the name of the target computing power service from the client includes: Receive the name of the target computing service sent by the Domain Name System (DNS) server, where the name of the target computing service is the domain name extracted by the DNS server from the domain name resolution request sent by the client.

24. The method according to claim 22, characterized in that After determining the address of the target computing power service, the method further includes: Determine the address of the target service instance corresponding to the address of the target computing service based on the pre-stored correspondence between the address of the computing service and the address of the service instance providing the computing service; Generate a second computing power routing table from the second gateway device to the target service node according to the address of the target service instance, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service; Send the second computing power routing table to the second gateway device.

25. The method according to claim 22, wherein Before obtaining the name of the target computing service from the client, the method further includes: Obtain static attribute information of the target service instance from the second gateway device, the static attribute information including the name of the target computing service; Allocate the address of the target computing power service to the target computing power service according to the static attribute information; Send the address of the target computing power service to the second gateway device.

26. The method according to claim 25, characterized in that The static attribute information also includes one or more of the address of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance and the capability of the target service instance.

27. The method according to any one of claims 22 to 26, characterized in that: The address of the computing power service and the address of the service instance are respectively in the same format as the network protocol IP address.

28. A computing power access method, characterized in that: Applied to a client, the client accesses a computing network through a first gateway device, and the method includes: Get the address of the target computing power service; Generate a first computing power access message with a destination address being an address of the target computing power service; A first computing power access message is sent to the first gateway device, and the client accesses the computing power network through the first gateway device, so that the first gateway device sends the first computing power access message to the target service node where the target service instance providing the target computing power service is located through the second gateway device, and the target service node accesses the computing power network through the second gateway device.

29. The method according to claim 28, characterized in that The step of obtaining the address of the target computing power service includes: Sending a domain name resolution request to a DNS server, where the domain name included in the domain name resolution request is the name of the target computing power service; Receive the address of the target computing service corresponding to the name of the target computing service fed back by the DNS server.

30. The method according to claim 28 or 29, characterized in that The address of the target computing power service is composed of a preset address prefix and the identifier of the target computing power service; In the computing power routing tables stored in the first gateway device and the second gateway device, the address of the target computing power service is represented by the identifier of the target computing power service.

31. The method according to claim 28 or 29, characterized in that The address of the target computing service and the address of the target service instance are respectively in the same format as the network protocol IP address.

32. A computing power access device, characterized in that: Applied to a first gateway device, the apparatus includes: A first receiving module is configured to receive a first computing power access message sent by a client, wherein the client accesses the computing power network through the first gateway device, and the destination address of the first computing power access message is the address of a target computing power service; A determination module is configured to determine an address of a second gateway device corresponding to the address of the target computing service based on a pre-stored first computing power routing table, wherein the first computing power routing table includes the address of the computing power service and the address of the gateway device to which the service node where the service instance providing the computing power service is located is connected when accessing the computing power network; an encapsulation module, configured to encapsulate the first computing power access message according to the address of the first gateway device and the address of the second gateway device to obtain a second computing power access message; A sending module is used to send the second computing power access message to the second gateway device.

33. The device according to claim 32, characterized in that The first computing power routing table also includes a routing cost for accessing a service node where a service instance providing computing power services is located; The determining module is specifically configured to determine the address of the second gateway device with the minimum routing cost corresponding to the address of the target computing power service according to a pre-stored first computing power routing table.

34. The device according to claim 32, characterized in that The encapsulation module is specifically used to encapsulate a segment routing SRv6 message header based on the sixth version of the Internet protocol in the outer layer of the first computing power access message to obtain a second computing power access message, where the source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next hop device of the first gateway device on the path from the first gateway device to the second gateway device.

35. The device according to claim 32, characterized in that The first computing power routing table also includes the address of the service instance providing computing power service; The determining module is further configured to determine the address of the target service instance corresponding to the address of the target computing service according to a pre-stored first computing power routing table; The encapsulation module is specifically used to encapsulate an SRv6 message header in the outer layer of the first computing power access message, and modify the destination address of the first computing power access message to the address of the target service instance to obtain a second computing power access message. The source address of the SRv6 message header is the address of the first gateway device, and the destination address of the SRv6 message header is the address of the next-hop device of the first gateway device on the path from the first gateway device to the second gateway device.

36. The device according to claim 35, characterized in that The SRv6 message header includes a segment routing header (SRH), and an optional field of the SRH is filled with an identifier indicating the address of the target computing power service.

37. The device according to claim 32, characterized in that The device further comprises: A second receiving module is configured to receive a routing protocol announcement sent by the second gateway device before receiving the first computing power access message sent by the client, wherein the routing protocol announcement includes static attribute information and dynamic attribute information of a target service instance that provides the target computing power service; A generation module is used to generate the first computing power routing table according to the static attribute information and the dynamic attribute information.

38. The device according to claim 37, characterized in that The static attribute information includes one or more of the address of the target computing service, the name of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance, and the capability of the target service instance; or The dynamic attribute information includes one or more of a load state and a session state of the target service instance.

39. The device according to any one of claims 32 to 38, characterized in that The address of the computing power service consists of a preset address prefix and an identifier of the computing power service; The address of the computing power service included in the first computing power routing table is represented by an identifier of the computing power service.

40. The device according to any one of claims 32 to 38, characterized in that The address of the computing power service and the address of the service instance are respectively in the same format as the network protocol IP address.

41. A computing power access device, characterized in that: Applied to a second gateway device, the apparatus includes: A first receiving module is configured to receive a second computing power access message from a first gateway device, where the first gateway device is a gateway device to which a client is connected when accessing a computing power network. The second computing power access message is a message obtained by encapsulating the first computing power access message according to the address of the first gateway device and the address of the second gateway device after the first gateway device determines the address of the second gateway device based on a pre-stored first computing power routing table and the destination address of the first computing power access message sent by the client. The destination address of the first computing power access message is the address of a target computing power service. The first computing power routing table includes the address of the computing power service and the address of the gateway device to which a service node where a service instance providing the computing power service is located is connected when accessing the computing power network. a decapsulation module, configured to decapsulate the second computing power access message to obtain a third computing power access message; a determination module, configured to determine an address of a target proxy device corresponding to the address of the target computing service based on a pre-stored second computing power routing table, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service; The first sending module is used to send the third computing power access message to the target service node where the target service instance providing the target computing power service is located according to the address of the target proxy device.

42. The device according to claim 41, characterized in that The second computing power access message includes a segment routing (SRv6) message header and a payload based on Internet Protocol version 6, where the source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is the first computing power access message; The decapsulation module is specifically configured to strip the SRv6 message header of the second computing power access message to obtain the first computing power access message, where the first computing power access message is the third computing power access message.

43. The device according to claim 41, characterized in that The first computing power routing table further includes an address of a service instance providing computing power services; the second computing power routing table further includes an address of a service instance providing computing power services; The second computing power access message includes an SRv6 message header and a payload, the source address of the SRv6 message header is the address of the first gateway device, the destination address of the SRv6 message header is the address of the second gateway device, and the payload is a third computing power access message obtained by modifying the destination address of the first computing power access message to the address of the target service instance; The decapsulation module is specifically configured to strip the SRv6 message header of the second computing power access message to obtain a third computing power access message; The determining module is specifically configured to determine the target service instance corresponding to the address of the target service instance according to the pre-stored second computing power routing table. The address of the proxy device.

44. The device according to claim 43, characterized in that The SRv6 message header includes a segment routing header (SRH), and an optional field of the SRH is filled with a target identifier indicating the address of the target computing power service; The determining module is specifically configured to determine, based on a pre-stored second computing power routing table, an address of a target proxy device corresponding to the target identifier and the address of the target service instance.

45. The device according to claim 41, characterized in that The device further comprises: A second receiving module is configured to receive the second computing power routing table issued by the operation and maintenance platform in the computing power network before receiving the second computing power access message; A storage module is used to store the second computing power routing table.

46. ​​The device according to claim 41, characterized in that The device further comprises: A third receiving module is configured to receive static attribute information and dynamic attribute information of the target service instance collected by the target proxy device before receiving the second computing power access message; A generation module is used to generate the second computing power routing table according to the static attribute information and the dynamic attribute information.

47. The device according to claim 46, characterized in that The dynamic attribute information includes one or more of a load state and a session state of the target service instance.

48. The device according to claim 41, characterized in that The device further comprises: a fourth receiving module, configured to receive static attribute information of the target service instance collected by the target proxy device before receiving the second computing power access message, wherein the static attribute information includes a name of the target computing power service; A second sending module is configured to send the static attribute information to an operation and maintenance platform in the computing power network if the second gateway device does not store the address of the target computing power service corresponding to the name of the target computing power service; The fifth receiving module is used to receive the address of the target computing power service allocated by the operation and maintenance platform to the target computing power service according to the static attribute information.

49. The device according to claim 48, characterized in that The device further comprises: The third sending module is used to send a routing protocol announcement to the first gateway device after obtaining the address of the target computing power service, where the routing protocol announcement includes static attribute information and dynamic attribute information of the target service instance.

50. The device according to any one of claims 46 to 49, characterized in that The static attribute information includes one or more of the address of the target computing service, the name of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance and the capabilities of the target service instance.

51. The device according to any one of claims 41 to 49, characterized in that The address of the computing power service consists of a preset address prefix and an identifier of the computing power service; The addresses of the computing power services included in the first computing power routing table and the second computing power routing table are both represented by identifiers of the computing power services.

52. The device according to any one of claims 41 to 49, characterized in that The address of the computing power service and the address of the service instance are respectively in the same format as the network protocol IP address.

53. A computing power access device, characterized in that: Applied to the operation and maintenance platform, the device includes: The first acquisition module is used to obtain the name of the target computing service from the client; A first determining module is configured to determine the address of a target computing service corresponding to the name of the target computing service based on a pre-stored correspondence between the name of the computing service and the address of the computing service; The first sending module is used to send the address of the target computing power service to the client, so that the client can access the target service node where the target service instance providing the target computing power service is located through the first gateway device and the second gateway device in the computing power network according to the address of the target computing power service. The client accesses the computing power network through the first gateway device, and the target service node accesses the computing power network through the second gateway device.

54. The device according to claim 53, characterized in that The first acquisition module is specifically used to receive the name of the target computing service sent by the domain name system DNS server, where the name of the target computing service is the domain name extracted by the DNS server from the domain name resolution request sent by the client.

55. The device according to claim 53, characterized in that The device further comprises: A second determining module is configured to, after determining the address of the target computing service, determine the address of the target service instance corresponding to the address of the target computing service based on a pre-stored correspondence between the address of the computing service and the address of the service instance providing the computing service; a generating module, configured to generate, based on the address of the target service instance, a second computing power routing table from the second gateway device to the target service node, wherein the second computing power routing table includes the address of the computing power service and the address of the proxy device of the service instance providing the computing power service; The second sending module is used to send the second computing power routing table to the second gateway device.

56. The device according to claim 53, characterized in that The device further comprises: A second acquisition module is configured to acquire static attribute information of the target service instance from the second gateway device before acquiring the name of the target computing service from the client, wherein the static attribute information includes the name of the target computing service; An allocation module, configured to allocate an address of the target computing power service to the target computing power service according to the static attribute information; The third sending module is used to send the address of the target computing power service to the second gateway device.

57. The device according to claim 56, characterized in that The static attribute information also includes one or more of the address of the target computing service, the service type of the target computing service, the address of the target service instance, the location of the target service instance, the priority of the target service instance and the capability of the target service instance.

58. The device according to any one of claims 53 to 57, characterized in that The address of the computing power service and the address of the service instance are respectively in the same format as the network protocol IP address.

59. A computing power access device, characterized in that: Applied to a client, the client accesses a computing network through a first gateway device, and the apparatus includes: The acquisition module is used to obtain the address of the target computing power service; A generating module, configured to generate a first computing power access message whose destination address is an address of the target computing power service; A sending module is used to send a first computing power access message to the first gateway device, and the client accesses the computing power network through the first gateway device, so that the first gateway device sends the first computing power access message to the target service node where the target service instance providing the target computing power service is located through the second gateway device, and the target service node accesses the computing power network through the second gateway device.

60. The device according to claim 59, characterized in that The acquisition module is specifically used to: Sending a domain name resolution request to a DNS server, where the domain name included in the domain name resolution request is the name of the target computing power service; Receive the address of the target computing service corresponding to the name of the target computing service fed back by the DNS server.

61. The device according to claim 59 or 60, characterized in that The address of the target computing power service is composed of a preset address prefix and the identifier of the target computing power service; In the computing power routing tables stored in the first gateway device and the second gateway device, the address of the target computing power service is represented by the identifier of the target computing power service.

62. The device according to claim 59 or 60, characterized in that The address of the target computing service and the address of the target service instance are respectively in the same format as the network protocol IP address.

63. A network device, characterized in that The method comprises a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions that can be executed by the processor, and the processor is prompted by the machine-executable instructions to: implement any of the methods described in claims 1-9, or implement any of the methods described in claims 10-21, or implement any of the methods described in claims 22-27, or implement any of the methods described in claims 28-31.

64. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 9, or implements the method described in any one of claims 10 to 21, or implements the method described in any one of claims 22 to 27, or implements the method described in any one of claims 28 to 31.

65. A computer program product, characterized in that When it is run on a computer, it enables the computer to execute the method of any one of claims 1 to 9, or execute the method of any one of claims 10 to 21, or execute the method of any one of claims 22 to 27, or execute the method of any one of claims 28 to 31.

Images