Data stream forwarding method and system, data routing system and data router

Abstract

The invention provides a data stream forwarding method and system, a data routing system and a data router, and the method comprises the steps: receiving a target data stream sent by a client based on a network entrance, and obtaining a service identifier through the target data stream; acquiring forwarding information corresponding to the service identifier based on the service identifier; and packaging the target data stream based on packaging information in the forwarding information to obtain a packaged data packet, and forwarding the packaged data packet to a network outlet based on a data forwarding path in the forwarding information. Through the method, a perfect data flow forwarding process is provided, the corresponding service node can be quickly matched, the matching time between the service request and the service node is shortened, the timeliness of service response is improved, and the efficiency and accuracy of data forwarding are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data stream forwarding in a communication process, and in particular to a data stream forwarding method, system, data routing system and data router. Background Art

[0002] Computing routing is an emerging data communication technology that aims to achieve efficient data forwarding and service scheduling by integrating computing and network resources. Traditional network routing is mainly based on network topology and path selection, focusing on the transmission efficiency of data packets and the load balancing of network traffic. However, with the development of technologies such as cloud computing, big data, and artificial intelligence, relying solely on network routing can no longer meet the needs of complex application scenarios. Computing routing technology came into being, combining the state perception of computing resources with network routing to achieve joint scheduling of computing and network resources.

[0003] The core idea of ​​computing power routing is to use the load, latency and other information of computing power nodes to dynamically select the best service path and guide data traffic to the optimal computing power node for processing. This process usually involves multiple key components, including computing power perception module, path selection module and traffic classification module. The computing power perception module is responsible for collecting and summarizing the status information of computing power nodes, such as CPU load, memory usage and network latency. The path selection module calculates the optimal forwarding path based on this information. The traffic classification module classifies the data flow and forwards it according to the predetermined strategy.

[0004] Currently, standardization organizations such as the Internet Engineering Task Force (IETF) and the China Communications Standards Association (CCSA) are actively promoting the standardization of computing power routing. For example, the IETF has established a computing-aware traffic selection (CATS) working group to focus on the demand and architecture research of computing power routing. The computing power routing system is usually composed of a computing network control center, computing power routers, and computing power nodes. The computing power router is the core network element, which includes core components such as CATS Traffic Classifier (C-TC), CATS Path Selector (C-PS), and CATS Forwarder (C-forwarder).

[0005] However, although IETF has defined the basic architecture and functions of computing power routers (C-TC, C-PS and C-forwarder), it has not detailed the specific data forwarding process and lacks clear operating procedures. Summary of the invention

[0006] In order to solve the problem of matching data forwarding paths and service nodes in a data service system, the present application provides a data stream forwarding method, system, data routing system and data router. Through the data stream forwarding method, the present application can achieve fast and accurate matching between data streams and service nodes.

[0007] In a first aspect, the present invention provides a data stream forwarding method, comprising:

[0008] Receive a target data stream sent by a client based on a network entry, and obtain a service identifier through the target data stream;

[0009] Acquire forwarding information corresponding to the service identifier based on the service identifier;

[0010] The target data flow is encapsulated based on the encapsulation information in the forwarding information to obtain an encapsulated data packet, and the encapsulated data packet is forwarded to a network egress based on the data forwarding path in the forwarding information.

[0011] Preferably, the method for obtaining the forwarding information includes:

[0012] Acquire a target service node corresponding to the service identifier based on the service identifier;

[0013] Receiving generation information of any service node based on a network entry to generate a service topology table, wherein the service identifier and a service address matching the service identifier are recorded in the service topology table, and the service address is used to access the corresponding service node;

[0014] When receiving the target data stream, searching the service topology table based on the service identifier to obtain the target service node;

[0015] The forwarding information corresponding to the target service node is obtained based on the target service node.

[0016] Preferably, the method for obtaining forwarding information further includes:

[0017] Acquire a target service node corresponding to the service identifier based on the service identifier;

[0018] Receiving a data routing table based on the network entry, and generating a data forwarding table based on the data routing table, wherein the data forwarding table records the service identifier and the data forwarding path and the encapsulation information that match the service identifier;

[0019] The data forwarding table is searched based on the service identifier of the target service node to obtain the forwarding information.

[0020] Preferably, based on the service identifier, a corresponding target service node is acquired;

[0021] When multiple target service nodes are acquired, an optimal service node is screened out according to the characteristic information of each target service node, and the corresponding forwarding information is obtained by using the optimal service node.

[0022] Preferably, based on the service identifier, a corresponding target service node is acquired;

[0023] When multiple target service nodes are acquired, the load is allocated to each target service node according to the weight parameter generated by the characteristic information of the target service node, and the service is transmitted between the multiple target service nodes to achieve load balancing among the multiple target service nodes.

[0024] Preferably, the characteristic information includes service metric information and network metric information, the service metric information is used to characterize the state and performance of service resources of the target service node, and the network metric information is used to characterize the state and performance of network resources of the network to which the target service node belongs;

[0025] A service metric table is also generated at the network entry, which records the service identifier and the service address of the service node matching the service identifier, the service metric information and the network metric information. The service address is used to access the corresponding service node.

[0026] Preferably, the target data stream carries the service identifier through an option parameter of the IP protocol or a DA address field, or the target data stream generates the service identifier through a flow feature at the network inlet.

[0027] Preferably, a service session table is also generated at the network entrance, wherein the service session table records the five-tuple information of the target data flow entering the network entrance in a historical period and the forwarding information corresponding to the five-tuple information;

[0028] When the target data flow can be matched to the service session table, the forwarding information is directly acquired based on the service session table.

[0029] Preferably, the service identifier includes an anycast address or a destination address of the TCP / IP protocol.

[0030] The present invention also provides a data routing system, which includes a control center, a data router and a plurality of service nodes, wherein the data router is arranged between the control center and the plurality of service nodes;

[0031] The data router includes a network entry connected to the control center for communication, a network exit connected to a plurality of the service nodes, and a functional component connected to the network entry and the network exit for communication;

[0032] The functional component is used to receive a target data stream sent by a client based on the network entry, and obtain a service identifier through the target data stream;

[0033] The functional component is also used to obtain corresponding forwarding information based on the service identifier, wherein the forwarding information includes a data forwarding path and encapsulation information;

[0034] The functional component is further configured to encapsulate the target data flow based on the encapsulation information in the forwarding information to obtain an encapsulated data packet, and forward the encapsulated data packet to a network egress based on the data forwarding path in the forwarding information.

[0035] Preferably, the functional component further acquires a target service node corresponding to the service identifier based on the service identifier;

[0036] The network entry is used to receive generation information from the control center to generate a service topology table, wherein the service identifier and a service address matching the service identifier are recorded in the service topology table, and the service address is used to access the corresponding target service node;

[0037] When receiving the target data flow, the functional component searches the service topology table based on the service identifier to obtain the target service node, and obtains the forwarding information corresponding to the target service node based on the target service node.

[0038] Preferably, the network entry is used to receive a data routing table from the control center, and generate a data forwarding table based on the data routing table, wherein the data forwarding table records the service identifier and the data forwarding path and the encapsulation information that match the service identifier;

[0039] The functional component also acquires a target service node corresponding to the service identifier based on the service identifier;

[0040] After acquiring the target service node, the functional component searches the data forwarding table based on the service identifier of the target service node to obtain the forwarding information.

[0041] Preferably, the functional component also obtains the target service node corresponding to the service identifier; when multiple target service nodes are obtained, the functional component also filters out the optimal service node according to the characteristic information of each target service node, and obtains the corresponding forwarding information with the optimal service node.

[0042] Preferably, the functional component also obtains the target service node corresponding to the service identifier based on the service identifier; when multiple target service nodes are obtained, the functional component also allocates load to each target service node according to a weight parameter generated according to characteristic information of each target service node, and performs service transmission between the multiple target service nodes to achieve load balancing between the multiple target service nodes.

[0043] Preferably, the characteristic information includes service metric information and network metric information, the service metric information is used to characterize the state and performance of service resources of the target service node, and the network metric information is used to characterize the state and performance of network resources of the network to which the target service node belongs;

[0044] The network entry is used to receive the characteristic information from the control center and generate a computing power measurement table. The service measurement table records the service identifier and the service address of the service node matching the service identifier, the service measurement information and the network measurement information. The service address is used to access the corresponding service node.

[0045] Preferably, the network portal is further used to receive information of the target data flow in a historical period and the related forwarding information from the control center to generate a service session table, wherein the service session table records quintuple information of the target data flow entering the network portal in a historical period and the forwarding information corresponding to the quintuple information;

[0046] When the target data flow can be matched to the service session table, the functional component directly obtains the forwarding information based on the service session table.

[0047] The present invention also provides a data router, which includes a functional component and a network inlet and a network outlet communicatively connected to the functional component, wherein the functional component works based on the data stream forwarding method as described above.

[0048] The present invention also provides a service data stream forwarding system, which comprises:

[0049] one or more processors;

[0050] Memory;

[0051] and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including instructions for executing the data stream forwarding method as described above.

[0052] The present invention also provides a computer-readable storage medium, which includes a computer program, and the computer program can be executed by a processor to implement the data stream forwarding method as described above.

[0053] The present invention also provides a computer program product, including a computer program, which implements the data stream forwarding method as described above when executed by a processor.

[0054] Compared with the prior art, the data stream forwarding method provided by the above technical solution of the present invention provides a complete data stream forwarding process from receiving the target data stream, obtaining the service identifier, to obtaining the forwarding information, encapsulating the data packet and forwarding it to the network exit. By carrying the service identifier in the target data stream, the corresponding forwarding information can be quickly matched, the matching time between the service request and the service node is shortened, the timeliness of the service response is improved, and the best forwarding path can be selected more accurately to ensure that the data stream is transmitted along the optimal path; in addition, by clarifying the data forwarding path and encapsulation information in the forwarding information, the encapsulation and forwarding process of the data packet is optimized, and the efficiency and accuracy of data forwarding are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 Schematic diagram of a data routing system in an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of a data stream forwarding method in an embodiment of the present invention.

[0057] Figure 3 The figure is a flow chart of a data stream forwarding method in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to explain the technical content, structural features, achieved objectives and effects of the present invention in detail, the following is a detailed description in conjunction with the implementation methods and the accompanying drawings.

[0059] This embodiment discloses a new data flow forwarding method for classified forwarding of data packets of a data router.

[0060] As a key network element of the computing power transmission network, the data router includes core functional components such as C-TC (traffic classifier), C-PS (path selector) and C-forwarder (traffic forwarder). Figure 1 ,The function of the data router is to forward the data packets from the client to the matching ,service node.

[0061] The data flow forwarding method in this embodiment controls the workflow of the data router having this architecture, such as Figure 1 and Figure 2 , the method comprises the following steps:

[0062] S1: receiving a target data stream sent by a client for requesting a service based on a network portal, and obtaining a service identifier through the target data stream, where the service identifier corresponds to a service node in the network that can provide the corresponding service;

[0063] S2: based on the service identifier, obtain the corresponding forwarding information, the forwarding information includes the data forwarding path and encapsulation information;

[0064] S3: Encapsulate the target data flow based on the encapsulation information in the forwarding information to obtain an encapsulated data packet, and forward the encapsulated data packet to a network egress based on the data forwarding path in the forwarding information.

[0065] After receiving the encapsulated data packet, the network egress removes the outer encapsulation and sends it to the corresponding service node.

[0066] In this embodiment, the network ingress is an ingress forwarding node of a data router, and the network egress is an egress forwarding node of a data router.

[0067] On the other hand, methods for obtaining forwarding information include:

[0068] Receiving generation information of any service node based on the network entrance to generate a service topology table, wherein the service identifier and a service address matching the service identifier are recorded in the service topology table, and the service address is used to access the corresponding service node;

[0069] When receiving the target data stream, searching the service topology table based on the service identifier in the target data stream to obtain the target service node;

[0070] Obtain corresponding forwarding information based on the target service node.

[0071] In this embodiment and the following embodiments, computing power service is used as an example for specific description, but the invention is not limited thereto.

[0072] When a new computing service (such as a cloud service node or a server that provides specific computing power) is launched, it needs to be registered and generated in the network. That is to say, for any service node, it must first register with the computing service network before it can provide computing services. Figure 3 , the control plane (e.g., the control center) receives the generation information of each service node and sends the generation information to the network portal, which generates and updates the service topology table based on the generation information. Then, on the data plane, when the network portal receives the target data stream, it can query the service topology table to quickly obtain the matching target service node.

[0073] Specifically, the generated information includes the service identifier (CS-ID), the service address (such as the IP address of the service node or the IP address of the API gateway of the data center) and the normalized computing power information (such as load percentage, delay parameters, etc.), then the generated service topology table is shown in Table 1. In addition, the service ID of the service node and the location address of the computing power service can be sent by extending the BGP update message, which can be the IP address of the service node or the IP address of the API gateway of the data center where the computing power service is located.

[0074] Table 1

[0075] CS-ID Service Address Hashrate Information 1001 Node A:192.168.1.1,... Load:30%,Latency:10ms 1001 Node B:192.168.1.2,... Load: 60%, Latency: 20ms

[0076] In this embodiment, the service topology table can be used to quickly and accurately match the required service nodes. Moreover, the control plane can update the generated information in real time, so that the newly registered service nodes can quickly enter the computing service network.

[0077] On the other hand, the method for obtaining forwarding information includes:

[0078] Obtain the corresponding target service node based on the service identifier;

[0079] Based on the network entry receiving data routing table, such as Figure 3 , and generate a data forwarding table based on the data routing table, the data forwarding table records the service identifier and the data forwarding path and encapsulation information matching the service identifier;

[0080] After the target service node is obtained, the data forwarding table is searched based on the service identifier of the target service node to obtain forwarding information.

[0081] In this embodiment, the data routing table can be generated by the management and control center (SMA) according to the collected network and service resource status information. The main function of the data routing table is to guide the path selection of the data flow in the network.

[0082] SMA sends the latest data routing table to the network portal in real time, and the network portal generates or updates the data forwarding table based on the data routing table.

[0083] The main function of the data forwarding table is to perform specific forwarding control of encapsulated data packets, and one data forwarding table is shown in Table 2 below.

[0084] Table 2

[0085] Instance ID Service Logo Traffic Classification Forwarding strategy Service segment identifier 1 CS-ID-1 1 SR-Policy1 END.DX1 1 CS-ID-2 1 SR-Policy2 END.DX2 1 CS-ID-3 1 SR-Policy3 END.DX3

[0086] In Table 2, Service Identifier (VRF): is used to distinguish different virtual routing instances. Each VRF represents an independent routing table, which can be used to support a multi-tenant environment or isolate different network services.

[0087] Service ID: used to uniquely identify a computing service.

[0088] Traffic Classification (TC): used to distinguish different types of traffic, which may be classified based on service priority, protocol type, etc. In Table 2, the TC of all entries is 1, indicating the same traffic classification level.

[0089] Forwarding policy: defines the forwarding path and encapsulation information of data packets in the network. For example, SR-Policy1 means that SRv6 (Segment Routing over IPv6) is used to encapsulate data packets in the IPv6 network, and the path information is attached to the data packet in the form of a label. It can be seen that the forwarding policy field not only defines the forwarding path of the data packet, but also includes various encapsulation information and policy control information required to implement the path. Through the forwarding policy, network devices can accurately control the transmission path of the data packet and perform corresponding encapsulation and forwarding operations according to the predefined policy.

[0090] Service segment identifier: used to identify the specific service segment in the forwarding strategy.

[0091] In this embodiment, by generating and updating the data forwarding table, the forwarding path and encapsulation information can be quickly matched, thereby further improving the efficiency and accuracy of data stream forwarding.

[0092] On the other hand, when multiple target service nodes are obtained, these multiple target service nodes can be enabled to enter the service process simultaneously, and the best service node can be screened out according to the characteristic information of each target service node, and the corresponding forwarding information can be obtained with the best service node.

[0093] When the multiple target service nodes are enabled to enter the service process at the same time, the load is allocated to each target service node according to the weight parameter generated by the characteristic information of the target service node, and the business is transmitted between the multiple target service nodes to achieve load balancing among the multiple target service nodes.

[0094] Furthermore, the characteristic information includes service metric information (S-Metric) and network metric information (N-Metric). The service metric information is used to characterize the status and performance of service resources of the target service node, and the network metric information is used to characterize the status and performance of network resources of the network to which the target service node belongs.

[0095] A service metric table is also generated at the network entrance, such as Figure 3The service metric table records the service identifier and the service address, service metric information and network metric information of the service node that matches the service identifier. The service address is used to access the corresponding service node.

[0096] In this embodiment, the computing power measurement is mainly used to measure the status and performance of service resources, which includes the following aspects:

[0097] Load: Indicates the current load level of the service node, usually expressed as a percentage. For example, a service node with a load of 70% means that it has used 70% of its computing power.

[0098] Computing power: represents the overall computing power of the service node, such as the number of processors, memory size, etc.

[0099] Response time: Indicates the speed at which a service node processes a request, usually expressed as a latency parameter. For example, the average response time of a service node is 50 milliseconds.

[0100] Availability: Indicates the availability status of the service node, such as whether it is under maintenance or has any faults.

[0101] Network metrics are mainly used to measure the status and performance of network resources. They include the following aspects:

[0102] Bandwidth: Indicates the available bandwidth of a network path. For example, the available bandwidth of a path is 100Mbps.

[0103] Latency: Indicates the average delay of a data packet through a network path, usually measured in milliseconds (ms). For example, the average latency of a path is 20 milliseconds.

[0104] Jitter: Indicates the delay variation of a network path, usually measured in milliseconds (ms).

[0105] Packet loss rate: Indicates the proportion of data packets lost in a network path, usually expressed as a percentage. For example, the packet loss rate of a path is 0.1%.

[0106] In this embodiment, through the configuration of the service metric table, as shown in Table 3 below, one or more optimal target service nodes can be screened out from the matched multiple target service nodes.

[0107] Table 3

[0108] CS-ID Service Address S-Metric N-Metric 1001 NodeA:192.168.1.1,... 0.7(70％) 0.3(30％) 1001 Node B:192.168.1.2,... 0.4(40％) 0.6(60％)

[0109] On the other hand, a service session table is also generated at the network entrance, as shown in Table 4 below. The service session table records the five-tuple information of the target data flow entering the network entrance in the historical period and the forwarding information corresponding to the five-tuple information;

[0110] When the target data flow can be matched to the service session table, the forwarding information is directly obtained based on the service session table.

[0111] Table 4

[0112]

[0113] In this embodiment, the service session table, such as Figure 3 , records the traffic information and corresponding forwarding information of the computing service request. When the target data flow does not request computing service for the first time through the data router, there is no need to query the service topology table and service metric table to obtain the forwarding information step by step. You only need to query the service session table to obtain the forwarding information recorded in the historical period. Ensure that the traffic of the same computing service request is always sent to the same service instance. In other words, the service session table helps ensure that the user's computing service request is always processed by the same service node, and the request will not be processed by different service nodes due to changes in the network, thereby ensuring the consistency and stability of the computing service.

[0114] On the other hand, the target data stream carries the service identifier through the option parameter of the IP protocol or the DA address field.

[0115] In addition, the service identifier includes an anycast address or a destination address of the TCP / IP protocol.

[0116] It should also be noted that after the network egress receives the encapsulated data packet, it selects the appropriate API gateway to forward the unpacked data packet to the corresponding service node.

[0117] There are two solutions for configuring the gateway at the network egress:

[0118] 1. The gateway at the network egress uses dual addresses (anycast address and interface address) to tunnel to the API gateway and forwards the message to the corresponding tunnel based on END.DX in the message.

[0119] 2. Each interface of the gateway at the network exit is routed to a computing power API gateway, and the message is forwarded to the corresponding interface based on END.DX in the message.

[0120] The gateway at the network exit refers to the gateway device at the end of the computing power routing path, which is responsible for directing the data flow to the service node.

[0121] For Solution 1, higher flexibility is provided, allowing one interface to connect to multiple computing power API gateways, but it needs to support dual addresses and tunneling technology.

[0122] For Solution 2, the network configuration is simplified, and each interface is connected to only one computing power API gateway, which is simpler to operate but less flexible.

[0123] In summary, the present invention discloses a data stream forwarding method. Figure 1 and Figure 3 , using a specific example to explain its workflow in detail.

[0124] For example, there is a video transcoding service, and user requests need to be forwarded to the best service node to ensure service quality. Assume that there are two service nodes (Node A and Node B), each providing different service capabilities and network status.

[0125] 1. Service Registration

[0126] When the video transcoding service goes online, the service will be registered on the control plane, and the network portal will generate a service topology table based on the generation information of the service on the network, as shown in Table 1 above.

[0127] 2. Client Request

[0128] The client sends a video transcoding request, assuming it is the first request, and the data packet carries CS-ID information (such as 1001).

[0129] 3. Network entry processing

[0130] The network entrance receives the data stream from the client and searches the service topology table based on the CS-ID. The result is: service information of Node A and Node B.

[0131] 4. Service measurement information search and path selection

[0132] The network portal searches for the relevant characteristic information of the service CS-ID in the service metric table (Table 3), selects the best service object, and selects Node A as the target service node.

[0133] 5. Data flow encapsulation and forwarding

[0134] According to the selected service identifier of Node A, the data forwarding table (Table 2) is searched to obtain the forwarding information of Node A including the forwarding path and encapsulation information, and the forwarding information is forwarded to the network egress.

[0135] 6. Network egress processing

[0136] The network egress receives the encapsulated data packet, removes the outer encapsulation, obtains the client data stream and forwards it to the API gateway of Node A.

[0137] 7. Data flow arrives at the service node

[0138] The data stream finally reaches Node A, completing the video transcoding service.

[0139] It should be noted that when the data stream forwarding of the above-mentioned video transcoding service is completed, the control plane will send the forwarding information related to the service to the network entry, and the network entry will maintain it in the service session table. In this way, when the target data stream corresponding to the video forwarding service enters the network exit again, the forwarding information can be directly matched through the service session table.

[0140] The data stream forwarding method disclosed in the above-mentioned embodiment of the present invention provides a complete data stream forwarding process from receiving the target data stream, obtaining the service identifier, searching the target service node, to obtaining the forwarding information, encapsulating the data packet and forwarding it to the network exit. By carrying the service identifier in the target data stream, the corresponding target service node can be quickly matched, the matching time between the service request and the service node is shortened, and the timeliness of the computing power service response is improved; in addition, by clarifying the data forwarding path and encapsulation information in the forwarding information, the encapsulation and forwarding process of the data packet is optimized, and the efficiency and accuracy of data forwarding are improved; furthermore, based on the forwarding information obtained from the target service node, the best forwarding path can be selected more accurately to ensure that the data stream is transmitted along the optimal path, thereby improving the service quality and user experience.

[0141] In another preferred embodiment of the present invention, a data routing system is also disclosed. Figure 1 , which includes a control center, a data router and several service nodes, and the data router is set between the control center and the several service nodes.

[0142] The data router includes a network entry connected to the control center, a network exit connected to several service nodes, and a functional component connected to the network entry and the network exit. The working components in this embodiment include C-TC, C-PS and C-forwarder.

[0143] The functional component is used to receive the target data stream sent by the client based on the network entrance, and obtain the service identifier through the target data stream;

[0144] The functional component is also used to obtain corresponding forwarding information based on the service identifier, and the forwarding information includes a data forwarding path and encapsulation information;

[0145] The functional component is also used to encapsulate the target data flow based on the encapsulation information in the forwarding information to obtain an encapsulated data packet, and forward the encapsulated data packet to a network exit based on the data forwarding path in the forwarding information.

[0146] The network egress is used to forward the received encapsulated data packets to the corresponding service node.

[0147] On the other hand, the functional component also obtains the target service node corresponding to the service identifier based on the service identifier;

[0148] The network entry is used to receive generation information from the control center to generate a service topology table. The service topology table records the service identifier and the service address that matches the service identifier. The service address is used to access the corresponding target service node.

[0149] When receiving the target data flow, the functional component searches the service topology table based on the service identifier to obtain the target service node, and obtains corresponding forwarding information based on the target service node.

[0150] On the other hand, the network entry is used to receive the data routing table from the control center and generate a data forwarding table based on the data routing table. The data forwarding table records the service identifier and the data forwarding path and encapsulation information matching the service identifier;

[0151] The functional component also obtains the target service node corresponding to the service identifier based on the service identifier;

[0152] After the target service node is obtained, the functional component searches the data forwarding table based on the service identifier of the target service node to obtain forwarding information.

[0153] On the other hand, the functional component also obtains the corresponding target service node based on the service identifier; when multiple target service nodes are obtained, the functional component also selects the optimal service node according to the characteristic information of each target service node, and obtains the corresponding forwarding information with the optimal service node.

[0154] On the other hand, the functional component also obtains the corresponding target service node based on the service identifier; when multiple target service nodes are obtained, the functional component also allocates load to each target service node according to the weight parameter generated by the characteristic information of each target service node, and performs service transmission between multiple target service nodes to achieve load balancing between multiple target service nodes.

[0155] Furthermore, the characteristic information includes service metric information and network metric information, the service metric information is used to characterize the state and performance of service resources of the target service node, and the network metric information is used to characterize the state and performance of network resources of the network to which the target service node belongs;

[0156] The network entry is used to receive feature information from the control center and generate a computing power measurement table. The service measurement table records the service identifier and the service address, service measurement information and network measurement information of the service node that matches the service identifier. The service address is used to access the corresponding service node.

[0157] On the other hand, the network entrance is also used to receive information about target data flows in historical periods and related forwarding information from the control center to generate a service session table, which records the five-tuple information of target data flows entering the network entrance in historical periods and the forwarding information corresponding to the five-tuple information;

[0158] When the target data flow can be matched to the service session table, the functional component directly obtains the forwarding information based on the service session table.

[0159] It should also be noted that the working principle and dynamic working process of the data routing system in this embodiment are detailed in the data stream forwarding method in the above embodiment, which will not be repeated here.

[0160] In another preferred embodiment of the present invention, a data router is disclosed, which includes a functional component and a network inlet and a network outlet communicatively connected to the functional component, and the functional component works based on the data stream forwarding method in the above embodiment.

[0161] The present invention also discloses a computing power service data stream forwarding system, which includes one or more processors, a memory and one or more programs, wherein one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the program includes instructions for executing the data stream forwarding method as described above. The processor can adopt a general central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits for executing related programs to implement the functions required to be performed by the modules in the computing power service data stream forwarding system of the embodiment of the present application, or to execute the data stream forwarding method of the method embodiment of the present application.

[0162] The present invention also discloses a computer-readable storage medium, which includes a computer program, and the computer program can be executed by a processor to complete the data stream forwarding method as described above. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a read-only memory (ROM), or a random access memory (RAM), or a magnetic medium, such as a floppy disk, a hard disk, a tape, a magnetic disk, or an optical medium, such as a digital versatile disc (DVD), or a semiconductor medium, such as a solid state disk (SSD), etc.

[0163] The embodiment of the present application also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of an electronic device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the electronic device executes the above-mentioned data stream forwarding method.

[0164] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. Therefore, equivalent changes made according to the scope of the patent application of the present invention are still within the scope covered by the present invention.

Claims

1. A data stream forwarding method, characterized in that: include: Receiving a target data stream sent by a client based on a network entry, and obtaining a service identifier through the target data stream; Acquire forwarding information corresponding to the service identifier based on the service identifier; The target data flow is encapsulated based on the encapsulation information in the forwarding information to obtain an encapsulated data packet, and the encapsulated data packet is forwarded to a network egress based on the data forwarding path in the forwarding information.

2. The data stream forwarding method according to claim 1, characterized in that: The method for obtaining the forwarding information includes: Acquire a target service node corresponding to the service identifier based on the service identifier; Receiving generation information of any service node based on a network entry to generate a service topology table, wherein the service identifier and a service address matching the service identifier are recorded in the service topology table, and the service address is used to access the corresponding service node; When receiving the target data stream, searching the service topology table based on the service identifier to obtain the target service node; The forwarding information corresponding to the target service node is obtained based on the target service node.

3. The data stream forwarding method according to claim 1, characterized in that: The method for obtaining forwarding information also includes: Acquire a target service node corresponding to the service identifier based on the service identifier; Receiving a data routing table based on the network entry, and generating a data forwarding table based on the data routing table, wherein the data forwarding table records the service identifier and the data forwarding path and the encapsulation information that match the service identifier; The data forwarding table is searched based on the service identifier of the target service node to obtain the forwarding information.

4. The data stream forwarding method according to claim 1, characterized in that: Acquire a target service node corresponding to the service identifier based on the service identifier; When multiple target service nodes are acquired, an optimal service node is screened out according to the characteristic information of each target service node, and the corresponding forwarding information is obtained by using the optimal service node.

5. The data stream forwarding method according to claim 1, characterized in that: Acquire a target service node corresponding to the service identifier based on the service identifier; When multiple target service nodes are acquired, the load is allocated to each target service node according to the weight parameter generated by the characteristic information of the target service node, and the service is transmitted between the multiple target service nodes to achieve load balancing among the multiple target service nodes.

6. The data stream forwarding method according to claim 4, characterized in that: The characteristic information includes service metric information and network metric information, wherein the service metric information is used to characterize the state and performance of service resources of the target service node, and the network metric information is used to characterize the state and performance of network resources of the network to which the target service node belongs; A service metric table is also generated at the network entry, which records the service identifier and the service address of the service node matching the service identifier, the service metric information and the network metric information. The service address is used to access the corresponding service node.

7. The data stream forwarding method according to claim 1, characterized in that: The target data stream carries the service identifier through an option parameter of the IP protocol or a DA address field, or the target data stream generates the service identifier through a flow feature at the network inlet.

8. The data stream forwarding method according to claim 1, characterized in that: A service session table is also generated at the network entrance, wherein the service session table records the five-tuple information of the target data flow entering the network entrance in a historical period and the forwarding information corresponding to the five-tuple information; When the target data flow can be matched to the service session table, the forwarding information is directly acquired based on the service session table.

9. The data stream forwarding method according to claim 1, characterized in that: The service identifier includes an anycast address or a destination address of the TCP / IP protocol.

10. A data routing system, characterized in that: It includes a control center, a data router and several service nodes, wherein the data router is arranged between the control center and the several service nodes; The data router includes a network entry connected to the control center for communication, a network exit connected to a plurality of the service nodes, and a functional component connected to the network entry and the network exit for communication; The functional component is used to receive a target data stream sent by a client based on the network entry, and obtain a service identifier through the target data stream; The functional component is also used to obtain corresponding forwarding information based on the service identifier, wherein the forwarding information includes a data forwarding path and encapsulation information; The functional component is further configured to encapsulate the target data flow based on the encapsulation information in the forwarding information to obtain an encapsulated data packet, and forward the encapsulated data packet to a network egress based on the data forwarding path in the forwarding information.

11. The data routing system according to claim 10, characterized in that: The functional component also acquires a target service node corresponding to the service identifier based on the service identifier; The network entry is used to receive generation information from the control center to generate a service topology table, wherein the service identifier and a service address matching the service identifier are recorded in the service topology table, and the service address is used to access the corresponding target service node; When receiving the target data flow, the functional component searches the service topology table based on the service identifier to obtain the target service node, and obtains the forwarding information corresponding to the target service node based on the target service node.

12. The data routing system according to claim 10, characterized in that: The network entry is used to receive a data routing table from the control center, and generate a data forwarding table based on the data routing table, wherein the data forwarding table records the service identifier and the data forwarding path and the encapsulation information that match the service identifier; The functional component also acquires a target service node corresponding to the service identifier based on the service identifier; After acquiring the target service node, the functional component searches the data forwarding table based on the service identifier of the target service node to obtain the forwarding information.

13. The data routing system according to claim 10, characterized in that: The functional component also obtains the corresponding target service node based on the service identifier; when multiple target service nodes are obtained, the functional component also filters out the optimal service node according to the characteristic information of each target service node, and obtains the corresponding forwarding information with the optimal service node.

14. The data routing system according to claim 10, characterized in that: The functional component also obtains the target service node corresponding to the service identifier based on the service identifier; when multiple target service nodes are obtained, the functional component also allocates load to each target service node according to the weight parameter generated by the characteristic information of each target service node, and performs service transmission between the multiple target service nodes to achieve load balancing between the multiple target service nodes.

15. The data routing system according to claim 13, characterized in that: The characteristic information includes service metric information and network metric information, wherein the service metric information is used to characterize the state and performance of service resources of the target service node, and the network metric information is used to characterize the state and performance of network resources of the network to which the target service node belongs; The network entry is used to receive the characteristic information from the control center and generate a computing power measurement table. The service measurement table records the service identifier and the service address of the service node matching the service identifier, the service measurement information and the network measurement information. The service address is used to access the corresponding service node.

16. The data routing system according to claim 13, characterized in that: The network entry is further used to receive information of the target data flow in a historical period and the related forwarding information from the control center to generate a service session table, wherein the service session table records quintuple information of the target data flow entering the network entry in a historical period and the forwarding information corresponding to the quintuple information; When the target data flow can be matched to the service session table, the functional component directly obtains the forwarding information based on the service session table.

17. A data router, characterized in that: It comprises a functional component and a network inlet and a network outlet communicatively connected to the functional component, and the functional component works based on the data stream forwarding method according to any one of claims 1 to 9.

18. A service data stream forwarding system, characterized in that: include: one or more processors; Memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs comprising instructions for executing the data stream forwarding method according to any one of claims 1 to 9.

19. A computer-readable storage medium, characterized in that: The method comprises a computer program, wherein the computer program can be executed by a processor to implement the data stream forwarding method according to any one of claims 1 to 9.

20. A computer program product, comprising a computer program, wherein when the program is executed by a processor, the data stream forwarding method according to any one of claims 1 to 9 is implemented.

Images