Network layer multipath forwarding method

Abstract

The invention provides a network layer multi-path forwarding method, and relates to a sending end, a receiving end and a network node. The method for processing the message by the network node comprises the following steps of: after receiving the message, determining whether the message has a multi-path forwarding requirement; obtaining a group of logic network topologies meeting application requirements, and determining a candidate logic topology set by combining the topology condition of the receiving end or the network node connected with the receiving end; determining a multi-path forwarding strategy according to the transmission intention field of the message header; selecting one topology from the candidate logic network topology set or still using the original topology based on a multi-path forwarding strategy; and when the logic network topology is selected, taking the network address of the receiving end as a transmission destination address, calculating through a routing algorithm to obtain a next forwarding node network address of the message, rewriting the destination address of the head of the message as the next forwarding node network address, and forwarding the data message. The method has the advantages that network resources are fully utilized, and diversified transmission requirements of users can be met.

Description

Technical Field

[0001] The present application belongs to the field of network technology and communication technology, and specifically relates to a network layer multi-path forwarding method. Background Art

[0002] The traditional TCP / IP protocol adopts a "best effort" single-path transmission method, which can no longer effectively meet users' needs for high-reliability and high-bandwidth transmission. As applications have higher and higher requirements for transmission rate and reliability, the single-path transmission method is facing problems such as low resource utilization, poor fault recovery capability and poor security.

[0003] In order to improve the efficiency and reliability of data transmission, multipath transmission technology has become an effective method and has been widely studied. Classic multipath transmission technology can be divided into: network layer multipath transmission technology and transport layer multipath transmission technology. Network layer multipath transmission technology forms multiple paths based on multipath routing protocols. The most classic one is the equal-cost multipath routing protocol, which achieves link load balancing to a certain extent, but is difficult to apply to asymmetric network topologies. In addition, since the utilization rate of each path is equal, it may increase the congestion of the congested path and waste the bandwidth resources of the idle path. Transport layer multipath transmission technology builds end-to-end multipaths by making full use of multiple network interfaces (with different network addresses) of network devices to achieve bandwidth aggregation and reliable transmission. MPTCP is a typical representative, but it is based on the transport layer. After selecting a pair of addresses, the transmission path in the network remains unchanged, and the idle bandwidth resources of network nodes cannot be fully utilized.

[0004] With the development of technologies such as big data, cloud computing, and AI, applications have put forward more diverse transmission requirements for the network, such as large bandwidth, low latency, high reliability, and high security. The existing multi-path research methods lack awareness of application needs and preferences, and it is difficult to support more flexible multi-path forwarding. Summary of the invention

[0005] The purpose of this application is to overcome the defects of the prior art, such as low transmission resource utilization, poor fault recovery capability and poor security.

[0006] In order to achieve the above-mentioned purpose, the present application proposes a network layer multi-path forwarding method, involving a sending end, a receiving end and a network node; a plurality of the network nodes form a network supporting the network layer multi-path forwarding method;

[0007] The network supports a plurality of logical network topologies that can meet different business requirements; the sending end, the receiving end and the network node are respectively located in one or more logical network topologies;

[0008] The method for processing a message by the network node includes:

[0009] Step 1: After receiving the message, match the multipath forwarding identification field in the message header to determine whether the message has a multipath forwarding requirement;

[0010] Step 2: Read the transmission requirement field in the message header and map it into a set of logical network topologies that meet the application requirements, and determine a candidate logical network topology set in combination with the topology of the receiving end or its connected network nodes;

[0011] Step 3: Determine a multipath forwarding strategy based on the transmission intention field in the packet header; select a topology from the candidate logical network topology set or use the original topology based on the multipath forwarding strategy;

[0012] Step 4: When the logical network topology is selected, the network address of the receiving end is used as the transmission destination address, the network address of the next forwarding node of the message is calculated through the routing algorithm, the destination address in the message header is rewritten as the network address of the next forwarding node, and the data message is forwarded; if the network address of the next forwarding node is the network address of the receiving end, the message is sent to the receiving end, and the network node completes the sending, otherwise go to step 1 and repeat.

[0013] As an improvement of the above method, the transmission requirement is used to represent the transmission service quality requirements put forward by users and applications to the network; the transmission service quality requirements include: bandwidth, packet loss rate, security, cost and delay;

[0014] The network maps the transmission requirements into a set of logical network topologies and corresponding routing algorithms that meet the transmission service quality requirements;

[0015] The fields in the data message header also include multipath forwarding parameters;

[0016] The mapping method is to maintain the mapping relationship based on the flow table, or to carry the available logical network topology in the multi-path forwarding parameters.

[0017] As an improvement of the above method, the transmission intention is used to express the intention requirements of users and applications when selecting network transmission resources, including: load balancing, bandwidth priority or latency priority; the network node reads this field and determines the multipath forwarding strategy based on this field.

[0018] As an improvement of the above method, the multipath forwarding identifier is used to identify whether the message uses multipath forwarding; the network node determines the multipath forwarding requirement of the message by reading and matching the multipath forwarding identifier.

[0019] As an improvement of the above method, the multipath forwarding parameters are used to help the network node determine a set of candidate logical network topologies and a topology transformation strategy; the multipath forwarding parameters include: an optional multipath forwarding topology set, topology information of the receiving end, and the remaining number of transformable topologies; the network node can read, fill in and modify this field.

[0020] As an improvement of the above method, the service requirements include: bandwidth, delay, packet loss or price.

[0021] As an improvement of the above method, the multi-path forwarding strategy is to select a suitable topology from a set of candidate logical network topologies or decide to still use the original topology;

[0022] The candidate logical network topology set is the intersection of the topology set corresponding to the transmission demand and the topology set where the receiving end or its connected network nodes are located;

[0023] The selection strategy for selecting a topology from the candidate logical network topology set includes: random selection, round-robin selection or weighted round-robin selection.

[0024] As an improvement of the above method, the method for the network node to obtain the network address of the receiving end includes: obtaining the network address of the receiving end through parsing by a parsing system based on the identification of the receiving end, or carrying the network address of the receiving end in an extension field of a message header;

[0025] The resolution system is a system for maintaining the mapping relationship between the network entity identifier and the corresponding network address.

[0026] As an improvement to the above method, the receiving end identifier is a unique and unchanged identifier used to represent the receiving end; the network can read but cannot modify this field.

[0027] As an improvement of the above method, the routing algorithm is a method for finding a forwarding path from a current node to a destination node in a logical network topology and calculating a network address of a next forwarding node; the routing algorithm includes: a shortest path and a DHT algorithm.

[0028] Compared with the prior art, the advantages of this application are:

[0029] A network layer multi-path forwarding method is proposed. Multiple forwarding paths are formed by changing and reselecting the logical network topology on the network nodes. User / application needs are perceived by carrying the demand and intention fields in the message header. The forwarding strategy corresponding to the demand / intention is used to fully utilize network resources to meet the goal of diversified transmission needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1Shown is a flow chart of the network layer multi-path forwarding method of the present application;

[0031] Figure 2 The figure shows the network layer multipath forwarding message structure of this application;

[0032] Figure 3 The figure shows a schematic diagram of the interaction of network layer multi-path forwarding entities of the present application;

[0033] Figure 4 Shown is a network layer multipath forwarding flow chart of an embodiment of the present application. DETAILED DESCRIPTION

[0034] The technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0035] In order to improve the elasticity and flexibility of data transmission, overlay routing technology can be used to provide customized transmission services. Overlay routing creates virtual logical links on top of the existing network to form a virtual logical topology. Routes under different logical topologies can form transmission paths that meet different user needs, which provides a new idea for multi-path transmission. However, how to perform multi-path forwarding based on a multi-topology network to meet the different transmission needs of users and applications is still a problem that needs to be solved.

[0036] The present application proposes a network layer multi-path forwarding method, which forms multiple forwarding paths by transforming and reselecting the logical network topology on the network nodes; based on the application requirements and intentions carried in the message, a forwarding strategy is selected to fully utilize network resources to meet the user's diverse transmission needs.

[0037] The network layer multipath forwarding method provided in the present application involves a data sending end, a receiving end and a network node, which constitute a network that supports the network layer multipath forwarding method; the network supports several logical network topologies that can meet different business needs, and the sending end, the receiving end and the network nodes are respectively located in one or more logical network topologies.

[0038] Logical network topology is a logical topology built based on some network nodes in the physical network and the physical and logical links between nodes. A physical network can form multiple logical topologies that meet different business requirements. Business requirements include: bandwidth, latency, packet loss, price, etc.

[0039] like Figure 1 As shown, the process of the network layer multipath forwarding method includes:

[0040] Step 1: The network node matches the multipath forwarding identifier: After receiving the data message, the network node matches the multipath forwarding identifier in the message header field to determine whether the message has a multipath forwarding requirement;

[0041] The data message header fields include: receiving end identification, transmission requirements, transmission intention, multi-path forwarding identification, and multi-path forwarding parameters.

[0042] The receiving end identifier is a unique and unchanged identifier used to represent the receiving end; the network can read but cannot modify this field.

[0043] Transmission requirements are used to represent the transmission service quality requirements that users and applications put forward for the network, including: bandwidth, packet loss rate, security, cost, latency, etc. The network can map transmission requirements into a set of logical network topologies and corresponding routing algorithms that meet the above requirements; mapping implementation methods include: maintaining mapping relationships based on flow tables, carrying available logical network topologies in multi-path forwarding parameters, etc.

[0044] Transmission intention is used to indicate the intention requirements of users and applications for network transmission resource selection, including load balancing, bandwidth priority, latency priority, etc. Network nodes can read this field and determine the multipath forwarding strategy based on it.

[0045] The multipath forwarding flag is used to identify whether a data message uses multipath forwarding. The network node determines the multipath forwarding requirement of the message by reading and matching this field.

[0046] The multipath forwarding parameters are used to help network nodes determine the candidate logical topology set and topology change strategy; the multipath forwarding parameters include: the optional multipath forwarding topology set, the topology information of the receiving end, the remaining number of topology changes, etc. The network node can read, fill in and modify this field.

[0047] Step 2: Determine the candidate logical topology set: The network node reads the transmission requirement field in the message header and maps it into a set of logical network topologies that can meet the application requirements. Combined with the topology of the receiving end or its connected network nodes, the candidate logical topology set is determined.

[0048] Step 3: Select a multi-path forwarding topology: The network node determines a multi-path forwarding strategy based on the transmission intention field in the message header, and selects a topology from the candidate logical topology set based on the strategy or continues to use the original topology;

[0049] The multipath forwarding strategy is used to select a suitable topology from a set of candidate logical network topologies or decide to continue using the original topology; the set of candidate logical topologies is the intersection of the topology set corresponding to the transmission demand and the topology set where the receiving end or its connected network nodes are located; the selection strategies include: random selection, polling selection, weighted polling selection, etc.

[0050] Step 4: Determine the network address of the next forwarding node: In the selected logical network topology, the network node uses the network address of the receiving end as the transmission destination address, calculates the network address of the next forwarding node of the data message through the routing algorithm, rewrites the destination address in the message header to the network address of the next forwarding node, and forwards the data message. If the network address of the next forwarding node is the network address of the receiving end, the message is sent to the receiving end to complete the sending, otherwise go to step 1 and repeat.

[0051] The method for a network node to obtain a receiving end network address includes: obtaining the receiving end network address through a resolution system based on the receiving end identifier and carrying the receiving end network address in an extension field of a message header. The resolution system is used to maintain a mapping relationship between a network entity identifier and a corresponding network address.

[0052] The routing algorithm is used to find the forwarding path from the current node to the destination node in a logical network topology and calculate the network address of the next forwarding node; the available routing algorithms include: shortest path, DHT algorithm, etc.

[0053] In order to make the technical solution of the present application clearer, the technical solution of the present application is described in detail below with reference to the accompanying drawings and embodiments.

[0054] In this embodiment, the network is divided into 8 logical topologies, which are numbered 1 to 8. The data transmitter UE1 is connected to the network node PE1, and the data receiver UE2 is connected to the network node PE2; the topology set where PE1 is located is {1, 2, 5}, and the topology set where PE2 is located is {2, 3, 5, 7}, and all topologies can be used for multipath forwarding; UE1 sends data to UE2.

[0055] like Figure 2 As shown, the embodiment of the present application provides a method for designing a message structure based on IPv6. In the embodiment, the multipath forwarding related fields are extended by using the IPv6 extension header. The definitions and values ​​of each field are as follows:

[0056] DID: 160 bits, receiving end identification;

[0057] MFlag: 1 bit, used to identify whether to use multipath forwarding, 1: enabled, 0: disabled.

[0058] Multipath forwarding parameters include:

[0059] REP: 4 bits, used to indicate the number of remaining replaceable topologies.

[0060] In this embodiment, the initial value REP=4

[0061] MPT1: 1 byte. This field is used to store information about participating in multipath topology. It uses a bitmap to represent the multipath topology that a node can select. The order of the topology number is the same as the order of the number of bits occupied.

[0062] In this embodiment, all topologies are optional, MPT1 = 0xFF (11111111)

[0063] MPT2: 1 byte, this field is used to store the topology information of the final destination node.

[0064] In this embodiment, PE2 is located in the topology set {2,3,5,7}, MPT2 = 0x69 (01101010)

[0065] TQoS: 1 byte, used to identify the transmission QoS requirements of the application for the network, including: bandwidth, packet loss rate, security, cost, and latency; can be filled in according to the type of application after application identification; TQoS is mapped to a set of logical topologies MPT3 that can meet the above requirements through the flow table.

[0066] In this embodiment, the mapping is: MPT3 = 0xFFFE (11111110)

[0067] TI: 1 byte, used to identify the intended requirements for selecting network transmission resources, including: bandwidth priority, price priority, and security priority.

[0068] In this implementation, TI is set to select bandwidth priority.

[0069] like Figure 3 As shown, the embodiment of the present application provides a schematic diagram of network layer multipath forwarding, Figure 4 This is a flowchart of network layer multipath forwarding in this embodiment. The specific steps of transmitting data from UE1 to UE2 are as follows:

[0070] Step 1) The sending end UE1 sends a message to the network node PE1 to which it is connected, and indicates the need to use multipath forwarding;

[0071] Step 2) After PE1 receives the message, it uses the IP extension header to add multipath forwarding related fields and fills them in, where: MFlag = 1, REP = 4, MPT1 = 0xFF, MPT2 = 0x69, and forwards the message;

[0072] Step 3) After receiving the message, the network node checks whether the MFlag flag is 1. If it is 1, go to step 4; otherwise, go to step 7);

[0073] Step 4) Use TQoS mapping to MPT3, and jointly determine the optional topology set TP with MPT1 and MPT2;

[0074] Step 5) Determine whether REP is greater than 1 and whether the number of optional topologies is greater than 1. If the conditions are met, go to step 6), otherwise go to step 7)

[0075] Step 6) Determine the matching strategy based on TI, and select a topology from the optional topology set as the forwarding topology of the message, REP--;

[0076] Step 7) Perform routing calculation in the current topology, obtain the next node, and forward the message;

[0077] Step 8) If the next forwarding node is PE2, go to step 9), otherwise go to step 3);

[0078] Step 9) PE2 removes the multipath forwarding related fields and forwards the message to UE2, and the process ends.

[0079] By transforming and reselecting the logical network topology on the network nodes, multiple forwarding paths are formed; and the forwarding strategy can be selected based on the application requirements and intentions carried in the message. The network layer multi-path forwarding method provided by this application achieves the ability to fully utilize network resources to meet the diverse transmission needs of users.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application is described in detail with reference to the embodiments, a person skilled in the art should understand that any modification or equivalent replacement of the technical solution of the present application does not depart from the spirit and scope of the technical solution of the present application and should be included in the scope of the claims of the present application.

Claims

1. A network layer multi-path forwarding method, involving a transmitting end, a receiving end and a network node; a plurality of the network nodes form a network supporting the network layer multi-path forwarding method; The network supports a plurality of logical network topologies that can meet different business requirements; the sending end, the receiving end and the network node are respectively located in one or more logical network topologies; It is characterized in that The method for processing a message by the network node includes: Step 1: After receiving the message, match the multipath forwarding identification field in the message header to determine whether the message has a multipath forwarding requirement; Step 2: Read the transmission requirement field in the message header and map it into a set of logical network topologies that meet the application requirements, and determine a candidate logical network topology set in combination with the topology of the receiving end or its connected network nodes; Step 3: Determine a multipath forwarding strategy based on the transmission intention field in the packet header; select a topology from the candidate logical network topology set or use the original topology based on the multipath forwarding strategy; Step 4: When the logical network topology is selected, the network address of the receiving end is used as the transmission destination address, the network address of the next forwarding node of the message is calculated through the routing algorithm, the destination address in the message header is rewritten as the network address of the next forwarding node, and the data message is forwarded; if the network address of the next forwarding node is the network address of the receiving end, the message is sent to the receiving end, and the network node completes the sending, otherwise go to step 1 and repeat.

2. The network layer multipath forwarding method according to claim 1, It is characterized in that The transmission requirement field is used to indicate the transmission service quality requirements put forward by users and applications to the network; The transmission service quality requirements include: bandwidth, packet loss rate, security, cost and delay; The network maps the transmission requirements into a set of logical network topologies and corresponding routing algorithms that meet the transmission service quality requirements; The fields in the data message header also include multipath forwarding parameters; The mapping method is to maintain the mapping relationship based on the flow table, or to carry the available logical network topology in the multi-path forwarding parameters.

3. The network layer multipath forwarding method according to claim 2, It is characterized in that The multipath forwarding parameters are used to help the network node determine a set of candidate logical network topologies and a topology transformation strategy; the multipath forwarding parameters include: an optional multipath forwarding topology set, topology information of the receiving end, and the number of remaining transformable topologies; the network node can read, fill in, and modify this field.

4. The network layer multipath forwarding method according to claim 1, It is characterized in that The transmission intention is used to indicate the intended requirements of users and applications for selecting network transmission resources, including: load balancing, bandwidth priority or delay priority; the network node reads this field and determines the multipath forwarding strategy based on this field.

5. The network layer multipath forwarding method according to claim 1, It is characterized in that The multipath forwarding identifier is used to identify whether the message uses multipath forwarding; the network node determines the multipath forwarding requirement of the message by reading and matching the multipath forwarding identifier.

6. The network layer multipath forwarding method according to claim 1, It is characterized in that The business requirements include: bandwidth, delay, packet loss or price.

7. The network layer multipath forwarding method according to claim 1, It is characterized in that The multipath forwarding strategy is to select a suitable topology from a set of candidate logical network topologies or decide to still use the original topology; The candidate logical network topology set is the intersection of the topology set corresponding to the transmission demand and the topology set where the receiving end or its connected network nodes are located; The selection strategy for selecting a topology from the candidate logical network topology set includes: random selection, round-robin selection or weighted round-robin selection.

8. The network layer multipath forwarding method according to claim 1, It is characterized in that The method for the network node to obtain the network address of the receiving end includes: obtaining it through parsing system based on the identification of the receiving end, or carrying the network address of the receiving end in the extension field of the message header; The resolution system is a system for maintaining the mapping relationship between the network entity identifier and the corresponding network address.

9. The network layer multipath forwarding method according to claim 8, It is characterized in that The receiving end identifier is a unique and unchanged identifier used to represent the receiving end; the network can read but cannot modify this field.

10. The network layer multipath forwarding method according to claim 1, It is characterized in that The routing algorithm is a method for finding a forwarding path from a current node to a destination node in a logical network topology and calculating the network address of the next forwarding node; The routing algorithms include: shortest path and DHT algorithm.

Images