SD-WAN network configuration method

Abstract

The invention discloses an SD-WAN network configuration method, an SD-WAN network comprises a controller, at least one POP and at least one CPE, the method is applied to the controller, and the method comprises the following steps: according to a control connection relationship between the controller and the POP in an original configuration tree, taking the controller as a root node of a new configuration tree, and taking the POP as a two-layer node of the new configuration tree; taking the CPE as a subordinate node of a corresponding POP in the new configuration tree according to a data connection relationship between the POP and the CPE in the original configuration tree; and taking the remaining CPE in the original configuration tree as a two-layer node of the new configuration tree, and completing conversion from a star-shaped control structure of the original configuration tree to a tree-shaped control structure of the new configuration tree. According to the method, the star topology of the SD-WAN control plane is converted into the tree topology, so that on one hand, the number of directly connected connections of the controller is reduced, the burden of the controller is reduced, and on the other hand, the communication quality of control connection is enhanced by using the transmission channel of the data plane.

Description

Technical Field

[0001] The present invention relates to the field of networking technology, and in particular to a SD-WAN network configuration method. Background Art

[0002] In the control plane of the relevant SD-WAN technology, all network nodes (such as CPE and POP) need to connect to the controller and receive the configuration sent by the controller. On the one hand, the SD-WAN network is relatively large and has many network nodes, so the controller needs to support a large number of control connections, which is a huge burden on the controller. On the other hand, the network nodes are distributed in various places, and the transmission quality of the control connection from the controller to each network node cannot be guaranteed. Summary of the invention

[0003] The technical problem to be solved by the present invention is that, in view of at least one defect of the related technology mentioned in the above background technology: there are many network nodes, and the controller needs to support a large number of control connections, which is a huge burden on the controller. In addition, the network nodes are distributed in various places, and the transmission quality of the control connection from the controller to each network node cannot be guaranteed. An SD-WAN network configuration method is provided.

[0004] The technical solution adopted by the present invention to solve the technical problem is: constructing a SD-WAN network configuration method, the SD-WAN network includes a controller, at least one POP and at least one CPE, and the method is applied to the controller, including the following steps:

[0005] S1: according to the control connection relationship between the controller and the POP in the original configuration tree, the controller is used as the root node of the new configuration tree, and the POP is used as the second-layer node of the new configuration tree;

[0006] According to the data connection relationship between the POP and the CPE in the original configuration tree, the CPE is used as a subordinate node of a corresponding POP in the new configuration tree;

[0007] S2: Use the remaining CPEs in the original configuration tree as the second-layer nodes of the new configuration tree to complete the conversion from the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

[0008] Optionally, in the SD-WAN network configuration method of the present invention, step S1 includes:

[0009] S11: according to the original configuration tree, the data connection relationship between the POP and the CPE, between the POP and the POP, and between the CPE and the CPE, as well as the control connection relationship between the controller and the POP and between the controller and the CPE and the connection overhead are put into the original set;

[0010] S12: putting the controller into a configuration tree set as the root node of the new configuration tree;

[0011] S13: in the original set, find a connection with the lowest cost whose one end is the controller and the other end is the POP, delete the found connection from the original set, and put the found POP into the configuration tree set as a second-layer node of the new configuration tree;

[0012] S14: Find a connection in the original set whose one end is the layer 2 node and the other end is the CPE, delete the found connection from the original set, and put the found CPE into the configuration tree set as a subordinate node of the layer 2 node;

[0013] S15: Determine whether there is a connection in the original set with one end being the controller and the other end being a POP, if so, return to step S13;

[0014] S16: If not, find all connections with POP in the original set, and delete the found connections from the original set.

[0015] Optionally, in the SD-WAN network configuration method of the present invention, step S11 further includes: according to the original configuration tree, putting the overhead of the connection between the POP and the CPE, between the POP and the POP, and between the CPE and the CPE into the original set;

[0016] Step S14 includes:

[0017] In the original set, find a connection whose one end is the layer 2 node and the other end is the CPE, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set;

[0018] If not, the found CPE is placed in the configuration tree set as a subordinate node of the layer 2 node;

[0019] If so, determine whether the cost of connecting the found CPE to the layer 2 node is less than the cost of the found CPE already hung on the layer 2 node. If so, put the found CPE into the configuration tree set as a subordinate node of the layer 2 node; if not, do not change the hung connection.

[0020] Optionally, in the SD-WAN network configuration method of the present invention, step S2 includes:

[0021] In the original set, find a connection with one end being the controller and the other end being the CPE, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set. If so, do not perform any operation; if not, put the found CPE into the configuration tree set as a layer-2 node of the new configuration tree, and complete the conversion from the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

[0022] Optionally, in the SD-WAN network configuration method of the present invention, step S2 further includes:

[0023] The connection with the found CPE in the original set is deleted, and the remaining connections in the original set are deleted.

[0024] Optionally, in the SD-WAN network configuration method of the present invention, the method further includes:

[0025] Receive registration requests sent by new POPs or CPEs after they come online for the first time;

[0026] The new POP or CPE is recorded according to the registration request, and configuration information is directly sent to the new POP or CPE, where the configuration information is used by the new POP or CPE to establish data connections with other POPs or CPEs.

[0027] Optionally, in the SD-WAN network configuration method of the present invention, the method further includes:

[0028] After detecting that a new CPE is connected to the SD-WAN network, checking whether there is a layer 2 node in the new configuration tree that has a data channel with the new CPE; if not, treating the new CPE as a layer 2 node directly connected to the controller;

[0029] If so, put the cost of connecting the new CPE to the Layer 2 node into the calculation set;

[0030] Taking out the layer 2 node with the minimum cost from the calculation set, and connecting the new CPE to the layer 2 node with the minimum cost as a new layer 3 node;

[0031] After the new configuration tree is adjusted, configuration information is sent to the new CPE according to the new configuration tree.

[0032] Optionally, in the SD-WAN network configuration method of the present invention, the method further includes:

[0033] After detecting that a new POP is connected to the SD-WAN network, check whether there is a layer 2 node in the new configuration tree that has a data channel with the new POP; if not, use the new POP as a layer 2 node directly connected to the controller;

[0034] If yes, add the cost of connecting the new POP to the layer 2 node and the cost of connecting to the controller into the calculation set;

[0035] A layer 2 node or controller with the minimum cost is selected from the calculation set, and a new POP is connected to the layer 2 node with the minimum cost as a new layer 3 node, or the new POP is used as a layer 2 node directly connected to the controller;

[0036] After the new configuration tree is adjusted, configuration information is sent to the new POP according to the new configuration tree.

[0037] Optionally, in the SD-WAN network configuration method of the present invention, the method further includes:

[0038] After detecting that the POP has left the SD-WAN network, determining whether the POP has a subordinate layer-3 node in the new configuration tree;

[0039] If so, after deleting the POP in the new configuration tree, reconnect the three-layer nodes under the original POP to the SD-WAN network, and adjust the new configuration tree according to the method described above;

[0040] If not, the POP and its connections are deleted in the new configuration tree.

[0041] Optionally, in the SD-WAN network configuration method of the present invention, the method further includes:

[0042] After detecting that the CPE leaves the SD-WAN network, determining whether the CPE is a layer 3 node in the new configuration tree;

[0043] If so, deleting the CPE and its connections in the new configuration tree;

[0044] If not, determine whether the CPE has any subordinate three-layer nodes in the new configuration tree. If so, after deleting the CPE in the new configuration tree, re-connect the three-layer nodes under the original CPE to the SD-WAN network, and adjust the new configuration tree according to the method described above; if not, delete the CPE and its connections in the new configuration tree.

[0045] By implementing the present invention, the following beneficial effects are achieved:

[0046] The present invention converts the star topology of the SD-WAN control plane into a tree topology, which, on the one hand, reduces the number of direct connections to the controller and reduces the burden on the controller, and on the other hand, utilizes the transmission channel of the data plane to enhance the communication quality of the control connection. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0048] Figure 1 A flow chart showing a method for configuring an SD-WAN network according to an embodiment of the present invention is shown;

[0049] Figure 2 A flowchart showing step S1 of the SD-WAN network configuration method according to an embodiment of the present invention is shown;

[0050] Figure 3 A schematic diagram showing an original configuration tree according to an embodiment of the present invention is shown;

[0051] Figure 4 Shows Figure 3 Schematic diagram of the controller and network nodes in the control plane;

[0052] Figure 5 Shows Figure 3 Schematic diagram of network nodes in the data plane;

[0053] Figure 6 Shown according to Figure 3 Schematic diagram of the original set generated, where e1 to e25 represent the overhead;

[0054] Figure 7 A schematic diagram showing a configuration tree set according to an embodiment of the present invention is shown;

[0055] Figure 8 , Fig.10 , Fig.12 , Fig.14 and Fig.16 A schematic diagram showing changes in the original set after repeatedly executing steps S13 and S14 in the SD-WAN network configuration method according to an embodiment of the present invention is shown;

[0056] Fig. 9 , Fig.11 , Fig.13 , Fig.15 and Fig.17 A schematic diagram showing changes in a configuration tree set after repeatedly executing steps S13 and S14 in an SD-WAN network configuration method according to an embodiment of the present invention is shown;

[0057] Fig.18 A schematic diagram showing an original set after executing step S16 in the SD-WAN network configuration method according to an embodiment of the present invention is shown;

[0058] Fig.19 A schematic diagram showing an original set after executing step S2 in the SD-WAN network configuration method according to an embodiment of the present invention is shown;

[0059] Fig. 20 A schematic diagram of a configuration tree set after executing step S2 in the SD-WAN network configuration method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0060] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0061] It should be noted that the flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor do they have to be executed in the order described. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0062] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0063] The following are explanations of the various terms:

[0064] The core idea of ​​SD-WAN (Software-Defined Wide Area Network) network technology is to manage and maintain each network node through centralized control (i.e. controller). As we all know, the Internet is an uncontrollable transmission, and the transmission quality of data on the Internet cannot be guaranteed. The SD-WAN network provides efficient and convenient management and maintenance for users to build cross-WAN networks by building an upper-layer controllable network based on the underlying Internet.

[0065] POP (Point-of-Presence), POP is the network service provider point of SD-WAN. SD-WAN network is an upper-layer controllable network based on the Internet, mobile network and some dedicated line resources built through POP. Simply put, POP is a network node at a geographical location. It contains a series of network devices and resources to ensure the reliability and performance of data transmission, such as WAN optimization devices, servers, storage devices, etc. In traditional networks, routers and switches can also be called POPs. In SD-WAN network, POP mainly refers to servers, balancers, WAN optimization devices, firewalls, etc.

[0066] CPE (Customer Premise Equipment), CPE is the terminal device for users to access the SD-WAN network. It is the dividing line between the SD-WAN network and the user's local network, and it is also the entrance to the SD-WAN network. The main functions for user traffic are implemented on the CPE network node. Simply put, CPE is a terminal device installed in the user's local network. CPE is mainly used to connect the user's internal network with the SD-WAN network, and to implement functions such as traffic management, security control, and network optimization. CPE is an important component on the user side. It can be a hardware device or a software solution. Common CPEs include SD-WAN access terminal devices, SD-WAN clients, etc.

[0067] Control plane: The nodes and connections that transmit control information are collectively referred to as the control plane. Figure 4 Schematic diagram of the controller and network nodes in the control plane.

[0068] Data plane: The nodes and connections that transmit user data are collectively referred to as the data plane. Figure 5 A schematic diagram of network nodes in the data plane is shown.

[0069] Some embodiments of the present invention disclose a SD-WAN network configuration method. The SD-WAN network includes a controller, at least one POP and at least one CPE. The method is applied to the controller. The controller controls the configuration and operation of the entire network. All configurations are generated by operations on the controller and finally sent to each CPE / POP device for execution. They take effect on the device to realize network functions.

[0070] The controller can be a controller that is being upgraded online or a new controller. Specifically, there are two scenarios where the conversion from a star control structure to a tree control structure is used. One is an online upgrade. The old version supports a star control structure, and the new version needs to switch seamlessly to a tree control structure. The other is a controller switch. The new controller that takes over the SD-WAN network does not have a tree control structure, and a new configuration tree needs to be generated based on the original configuration tree. It should be noted that the configuration tree is an overall logical concept. The controller needs to record the overall configuration tree, and each network node (POP and CPE) also needs to record the path from the node to the controller.

[0071] like Figure 1 As shown, the SD-WAN network configuration method includes the following steps:

[0072] S1: According to the control connection relationship between the controller and POP in the original configuration tree, the controller is used as the root node of the new configuration tree, and the POP is used as the second-layer node of the new configuration tree;

[0073] According to the data connection relationship between the POP and the CPE in the original configuration tree, the CPE is used as a subordinate node of a corresponding POP in the new configuration tree;

[0074] S2: The remaining CPEs in the original configuration tree are used as the second-layer nodes of the new configuration tree to complete the conversion from the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

[0075] In some embodiments, such as Figure 2 As shown, step S1 includes:

[0076] S11: according to the original configuration tree, data connection relationships between POPs and CPEs, between POPs and POPs, and between CPEs and CPEs, as well as control connection relationships between controllers and POPs and between controllers and CPEs and connection overheads are put into the original set;

[0077] S12: putting the controller into the configuration tree set as the root node of the new configuration tree;

[0078] S13: in the original set, find a connection with the lowest cost, one end of which is a controller and the other end is a POP, delete the found connection from the original set, and put the found POP into the configuration tree set as a second-layer node of the new configuration tree;

[0079] S14: in the original set, find a connection whose one end is the layer 2 node and the other end is the CPE, delete the found connection from the original set, and put the found CPE into the configuration tree set as a subordinate node of the layer 2 node;

[0080] S15: Determine whether there is a connection in the original set with one end being a controller and the other end being a POP. If so, return to step S13;

[0081] S16: If not, find all connections with POP in the original set, and delete the found connections from the original set.

[0082] In some embodiments, step S11 further includes: according to the original configuration tree, putting the overhead of the connection between the POP and the CPE, between the POP and the POP, and between the CPE and the CPE into the original set;

[0083] Step S14 includes:

[0084] In the original set, find a connection whose one end is the layer 2 node and the other end is the CPE, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set;

[0085] If not, the found CPE is placed in the configuration tree set as a subordinate node of the layer 2 node;

[0086] If so, determine whether the cost of the CPE found to be connected to the Layer 2 node is less than the cost of the CPE found to be already attached to the Layer 2 node. If so, put the found CPE into the configuration tree set as a subordinate node of the Layer 2 node; if not, do not change the already attached connection.

[0087] Among them, overhead refers to information such as network delay, packet loss, jitter, and bandwidth consumption. Connection overhead is generally used to evaluate connection quality.

[0088] In some embodiments, step S2 includes:

[0089] In the original set, find the connection with one end as the controller and the other end as the CPE, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set. If so, do not do anything; if not, put the found CPE into the configuration tree set as a second-layer node of the new configuration tree, completing the conversion from the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

[0090] In some embodiments, step S2 includes:

[0091] In the original set, find a connection with a controller at one end and a CPE at the other end, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set. If so, do not do anything; if not, put the found CPE into the configuration tree set as a second-layer node of the new configuration tree.

[0092] The connection with the found CPE in the original set is deleted, and the remaining connections in the original set are deleted, completing the conversion of the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

[0093] The specific implementation is as follows:

[0094] Step 1: According to Figure 3 The original configuration tree shown in FIG. 1 is a diagram showing the data connection relationship and connection overhead between POP and CPE, between POP and POP, and between CPE and CPE, as well as the control connection relationship and connection overhead between the controller and POP and between the controller and CPE. Figure 6 Put the controller into the original collection as shown. Figure 7 In the configuration tree collection shown, it serves as the root node of the new configuration tree.

[0095] Step 2: If Figure 8 As shown, in the original set, find the connection with the lowest cost, one end of which is the controller and the other end is POP, and delete the found connection (such as controller---POP1) from the original set. Fig. 9As shown, the found POP (such as POP1) is put into the configuration tree set as a second-level node of the new configuration tree.

[0096] Step 3: If Fig.10 As shown, in the original set, find a connection with one end being the layer 2 node (such as POP1) and the other end being the CPE, delete the found connection (such as CPE1-POP1) from the original set, and determine whether the found CPE (such as CPE1) already exists in the configuration tree set;

[0097] If not, then Fig.11 As shown, the found CPE (such as CPE1) is placed in the configuration tree set as a subordinate node of the second-layer node (such as POP1);

[0098] If so, determine whether the cost of the found CPE (such as CPE1) connected to the second-layer node (such as POP1) is less than the cost of the found CPE (such as CPE1) already hung on the second-layer node. If so, put the found CPE (such as CPE1) into the configuration tree set as a subordinate node of the second-layer node; if not, do not change the hung connection.

[0099] Step 4: Determine whether there is a connection in the original set with one end being a controller and the other end being a POP. If so, return to repeat the second and third steps, as shown in Fig.12 , Fig.13 , Fig.14 , Fig.15 , Fig.16 and Fig.17 As shown, Fig.16 and Fig.17 The found CPE5 shown in already exists in the configuration tree set. At this time, it is determined that the cost of CPE5 connecting to POP4 is less than the cost of CPE5 hanging on POP2, so CPE5 is used as a subordinate node of POP4. If not, then Fig.18 As shown, all connections with POP are found in the original set, and the found connections are deleted from the original set.

[0100] Step 5: If Fig.19 As shown, in the original set, find a connection with a controller at one end and a CPE at the other end, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set. If so, do not do anything; if not, Fig. 20 As shown, the found CPE (such as CPE7) is placed in the configuration tree set as a second-layer node of the new configuration tree. Fig.19 As shown, the connection with the found CPE in the original set (such as CPE6-CPE7) is deleted.

[0101] Step 6: Delete the remaining connections in the original set (such as CPE1-CPE2, CPE3-CPE4), and complete the following steps: Figure 4 The star control structure of the original configuration tree shown in Fig. 20 The new configuration tree shown is a transformation of the tree control structure.

[0102] The final new configuration tree is three-layer, in which the root node must be the controller; the second-layer node is generally a POP, and may also be a CPE (the CPE does not have a data channel with any POP); the third-layer node is the CPE.

[0103] In some embodiments, the SD-WAN network configuration method further includes:

[0104] Receive registration requests sent by new POPs or CPEs after they come online for the first time;

[0105] The new POP or CPE is recorded according to the registration request, and the configuration information is directly sent to the new POP or CPE. The configuration information is used by the new POP or CPE to establish a data connection with other POP or CPE.

[0106] In some embodiments, the SD-WAN network configuration method further includes:

[0107] After detecting that a new CPE is connected to the SD-WAN network, check whether there is a Layer 2 node in the new configuration tree that has a data channel with the new CPE; if not, treat the new CPE as a Layer 2 node directly connected to the controller;

[0108] If so, put the cost of connecting the new CPE to the Layer 2 node into the calculation set;

[0109] The Layer 2 node with the minimum cost is selected from the calculation set, and the new CPE is connected to the Layer 2 node with the minimum cost as a new Layer 3 node.

[0110] After the new configuration tree is adjusted, the configuration information is sent to the new CPE according to the new configuration tree. When sending configuration information, the controller searches for the configuration path according to the new configuration tree. If it is directly connected, it is sent directly. If it is forwarded through a layer 2 node, it is packaged and sent to the layer 2 node, and then the layer 2 node transmits the configuration to the final node through the data channel.

[0111] Among them, check whether there is a second-layer node in the new configuration tree that has a data channel with the new CPE. Specifically, the controller checks it through the configuration information directly sent down. For example, the user has three branches online and wants to add a new branch. The user requires the new branch to interact with two of the three branches. The controller will directly send the configuration information to the user to establish a channel between the new branch and the other two branches.

[0112] In some embodiments, the SD-WAN network configuration method further includes:

[0113] After detecting that a new POP is connected to the SD-WAN network, check whether there is a Layer 2 node in the new configuration tree that has a data channel with the new POP; if not, treat the new POP as a Layer 2 node directly connected to the controller;

[0114] If yes, put the cost of connecting the new POP to the Layer 2 node and the cost of connecting to the controller into the calculation set;

[0115] Select the Layer 2 node with the minimum cost in the calculation set, connect the new POP to the Layer 2 node with the minimum cost as a new Layer 3 node, or select the controller with the minimum cost in the calculation set and use the new POP as a Layer 2 node directly connected to the controller.

[0116] After the new configuration tree is adjusted, the configuration information is sent to the new POP according to the new configuration tree. When sending configuration information, the controller searches for the configuration path according to the new configuration tree. If it is directly connected, it is sent directly. If it is forwarded through a second-layer node, it is packaged and sent to the second-layer node, and then the second-layer node transmits the configuration to the final node through the data channel.

[0117] It should be noted that, by default, POP is prioritized as a Layer 2 node, and CPE is directly connected to the controller as a Layer 2 node only when it is not connected to POP.

[0118] In some embodiments, the SD-WAN network configuration method further includes:

[0119] After detecting that the POP has left the SD-WAN network, determine whether the POP has subordinate Layer 3 nodes (i.e., CPE) in the new configuration tree;

[0120] If so, after deleting the POP in the new configuration tree, reconnect the Layer 3 nodes under the original POP to the SD-WAN network, and adjust the new configuration tree according to the above-mentioned method of connecting the new CPE to the SD-WAN network;

[0121] If not, delete the POP and its connections in the new configuration tree.

[0122] In some embodiments, the SD-WAN network configuration method further includes:

[0123] After detecting that the CPE has left the SD-WAN network, determine whether the CPE is a Layer 3 node in the new configuration tree;

[0124] If so, delete the CPE and its connections in the new configuration tree;

[0125] If not, determine whether the CPE has any subordinate three-layer nodes in the new configuration tree. If so, after deleting the CPE in the new configuration tree, re-connect the original three-layer nodes under the CPE to the SD-WAN network, and adjust the new configuration tree according to the above-mentioned method of new CPE access to the SD-WAN network; if not, delete the CPE and its connection in the new configuration tree.

[0126] By implementing the present invention, the following beneficial effects are achieved:

[0127] The present invention converts the star topology of the SD-WAN control plane into a tree topology, which, on the one hand, reduces the number of direct connections to the controller and reduces the burden on the controller, and on the other hand, utilizes the transmission channel of the data plane to enhance the communication quality of the control connection.

[0128] Specifically, in the solution of the present invention, since a large number of nodes are configured through other nodes, the number of connections directly connected to the controller is greatly reduced, which reduces the burden on the controller. For example, there are 300 nodes in an SD-WAN network. Under normal circumstances, each node needs to have a connection to the controller, that is, 300 network connections. If 50 of them are used as layer 2 nodes and the rest as layer 3 nodes in advance using the solution of the present invention, the number of network connections that the controller needs to maintain is reduced to 50.

[0129] Generally speaking, the second-layer nodes are mainly POPs, and POPs and controllers are generally deployed in the cloud, which are relatively close and have good link quality. The data of the third-layer nodes is forwarded through the second-layer nodes, and the data is encapsulated in the data channel of the data plane. The data plane uses various technologies to improve the link quality to a certain extent (such as using multiple transmission and selection, TCP acceleration, QoS, routing and other technologies to improve the transmission quality of the data channel). Therefore, sending configurations to the third-layer nodes in this way will have better transmission quality than direct connection.

[0130] It can be understood that the above embodiments only express some implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the patent scope of the present invention. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present invention, the above embodiments or technical features can be freely combined, and several deformations and improvements can be made, which all belong to the protection scope of the present invention, that is, the embodiments described in "some embodiments" can be freely combined with any of the above and below embodiments; therefore, all equivalent changes and modifications made to the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A SD-WAN network configuration method, characterized in that: The SD-WAN network includes a controller, at least one POP and at least one CPE, and the method is applied to the controller, comprising the following steps: S1: according to the control connection relationship between the controller and the POP in the original configuration tree, the controller is used as the root node of the new configuration tree, and the POP is used as the second-layer node of the new configuration tree; According to the data connection relationship between the POP and the CPE in the original configuration tree, the CPE is used as a subordinate node of a corresponding POP in the new configuration tree; S2: Use the remaining CPEs in the original configuration tree as the second-layer nodes of the new configuration tree to complete the conversion from the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

2. The SD-WAN network configuration method according to claim 1, characterized in that: Step S1 includes: S11: according to the original configuration tree, the data connection relationship between the POP and the CPE, between the POP and the POP, and between the CPE and the CPE, as well as the control connection relationship between the controller and the POP and between the controller and the CPE and the connection overhead are put into the original set; S12: putting the controller into a configuration tree set as the root node of the new configuration tree; S13: in the original set, find a connection with the lowest cost whose one end is the controller and the other end is the POP, delete the found connection from the original set, and put the found POP into the configuration tree set as a second-layer node of the new configuration tree; S14: Find a connection in the original set whose one end is the layer 2 node and the other end is the CPE, delete the found connection from the original set, and put the found CPE into the configuration tree set as a subordinate node of the layer 2 node; S15: Determine whether there is a connection in the original set with one end being the controller and the other end being a POP, if so, return to step S13; S16: If not, find all connections with POP in the original set, and delete the found connections from the original set.

3. The SD-WAN network configuration method according to claim 2, characterized in that: Step S11 also includes: according to the original configuration tree, putting the overhead of the connection between POP and CPE, between POP and POP, and between CPE and CPE into the original set; Step S14 includes: In the original set, find a connection whose one end is the layer 2 node and the other end is the CPE, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set; If not, the found CPE is placed in the configuration tree set as a subordinate node of the layer 2 node; If so, determine whether the cost of connecting the found CPE to the layer 2 node is less than the cost of the found CPE already hung on the layer 2 node. If so, put the found CPE into the configuration tree set as a subordinate node of the layer 2 node; if not, do not change the hung connection.

4. The SD-WAN network configuration method according to claim 2, characterized in that: Step S2 includes: In the original set, find a connection with one end being the controller and the other end being the CPE, delete the found connection from the original set, and determine whether the found CPE already exists in the configuration tree set. If so, do not perform any operation; if not, put the found CPE into the configuration tree set as a layer-2 node of the new configuration tree, and complete the conversion from the star control structure of the original configuration tree to the tree control structure of the new configuration tree.

5. The SD-WAN network configuration method according to claim 4, characterized in that: Step S2 also includes: The connection with the found CPE in the original set is deleted, and the remaining connections in the original set are deleted.

6. The SD-WAN network configuration method according to claim 1, characterized in that: The method further comprises: Receive registration requests sent by new POPs or CPEs after they come online for the first time; The new POP or CPE is recorded according to the registration request, and configuration information is directly sent to the new POP or CPE, where the configuration information is used by the new POP or CPE to establish data connections with other POPs or CPEs.

7. The SD-WAN network configuration method according to claim 1 or 6, characterized in that: The method further comprises: After detecting that a new CPE is connected to the SD-WAN network, checking whether there is a layer 2 node in the new configuration tree that has a data channel with the new CPE; if not, treating the new CPE as a layer 2 node directly connected to the controller; If so, put the cost of connecting the new CPE to the Layer 2 node into the calculation set; Taking out the layer 2 node with the minimum cost from the calculation set, and connecting the new CPE to the layer 2 node with the minimum cost as a new layer 3 node; After the new configuration tree is adjusted, configuration information is sent to the new CPE according to the new configuration tree.

8. The SD-WAN network configuration method according to claim 1 or 6, characterized in that: The method further comprises: After detecting that a new POP is connected to the SD-WAN network, check whether there is a layer 2 node in the new configuration tree that has a data channel with the new POP; if not, use the new POP as a layer 2 node directly connected to the controller; If yes, add the cost of connecting the new POP to the layer 2 node and the cost of connecting to the controller into the calculation set; A layer 2 node or controller with the minimum cost is selected from the calculation set, and a new POP is connected to the layer 2 node with the minimum cost as a new layer 3 node, or the new POP is used as a layer 2 node directly connected to the controller; After the new configuration tree is adjusted, configuration information is sent to the new POP according to the new configuration tree.

9. The SD-WAN network configuration method according to claim 7, characterized in that: The method further comprises: After detecting that the POP has left the SD-WAN network, determining whether the POP has a subordinate layer-3 node in the new configuration tree; If so, after deleting the POP in the new configuration tree, reconnect the three-layer nodes under the original POP to the SD-WAN network, and adjust the new configuration tree according to the method of claim 7; If not, the POP and its connections are deleted in the new configuration tree.

10. The SD-WAN network configuration method according to claim 7, characterized in that: The method further comprises: After detecting that the CPE leaves the SD-WAN network, determining whether the CPE is a layer 3 node in the new configuration tree; If so, deleting the CPE and its connections in the new configuration tree; If not, determine whether the CPE has any subordinate three-layer nodes in the new configuration tree. If so, after deleting the CPE in the new configuration tree, re-connect the three-layer nodes subordinate to the original CPE to the SD-WAN network, and adjust the new configuration tree according to the method described in claim 7; if not, delete the CPE and its connections in the new configuration tree.

Images