WO2015035598A1

WO2015035598A1 - Method and apparatus for adjusting link overhead

Info

Publication number: WO2015035598A1
Application number: PCT/CN2013/083444
Authority: WO
Inventors: 张旭东; 胡志波
Original assignee: 华为技术有限公司
Priority date: 2013-09-13
Filing date: 2013-09-13
Publication date: 2015-03-19
Also published as: CN105247823A; CN105247823B

Abstract

Embodiments of the present invention provide a method for adjusting a link overhead and can prevent occurrence of a micro ring phenomenon. The method comprises: determining at least one pair of to-be-processed nodes; determining a path overhead change value of each to-be-processed node, wherein the path overhead change value of the each to-be-processed node is a difference between a first path overhead and a second path overhead of the each to-be-processed node; based on the path overhead change value of the to-be-processed nodes, performing at least two times of adjustment on a link overhead of a first link, and when the first link is recovered from a fault, making a next-hop node in a first path in each pair of to-be-processed nodes migrate an optimal path to a first node to the first path earlier than a previous-hop node, or when the first link is faulty, making the previous-hop node in the first path in the each pair of to-be-processed nodes migrate the optimal path to the first node out of the first path earlier than the next-hop node.

Description

Method and device for adjusting link overhead

The present invention relates to the field of communications and, more particularly, to a method and apparatus for adjusting link overhead. Background technique

At present, a technology is known in which a network gateway device (IP, Internet Protocol) network network device uses a distributed intra-gateway protocol (Intercomm Gateway Protocol) calculation, that is, each network device separately performs routing calculations, and each other There is no order control between them. Therefore, when a link fault or fault recovery occurs, the start time, run time, end time of the calculation, and delivery of the route calculation are caused by the difference in the hardware capabilities of the network devices and the difference between the internal and external environments of the device. The time points of the forwarding information base (FIB, Fowarding Information Base) entries are inconsistent, which leads to the occurrence of microrings.

Specifically, in the IP network shown in FIG. 2, for example, when the link between the node a and the node b recovers from a failure, for example, if the hardware capability of the node c is stronger than the hardware capability of the node b, and If the load of the node c is lower than the load of the node b, the calculation of the route by the node c and the issuance of the FIB entry are completed before the node b, and further, the optimal path of the node c from the node c to the node a is from the path 1 (The path is node c, node d, node e, node f, node a in turn in the forwarding order) to path 2 (the path includes node c, node b, node a in turn in the forwarding order), or node c Switching from link c→d to link c→b, node c sends the data to node b when it needs to send data to node a. At this time, the calculation of the route by the node b and the delivery of the FIB entry are not completed yet, and the node b still has the path 3 (the path includes the node b, the node c, the node d, the node e, the node f, and the node in the order of forwarding. a) as the optimal path from node b to node a, therefore, when node b needs to send data to node a, select link 1) → (, send the data to node c. Thus, at node b and node A micro-ring phenomenon occurs between c, that is, data is forwarded multiple times between node b and node c, which may cause data to be transmitted in the IP network for too long and discarded.

Therefore, it is desirable to provide a technique capable of preventing the occurrence of a microring phenomenon. Summary of the invention

The present invention provides a method and apparatus for adjusting link overhead, which can prevent micro ring phenomenon Now.

In a first aspect, a method for adjusting link overhead is provided, which is implemented in a communication system including at least three nodes, wherein a first node and a second node are directly connected by a first link, the first chain The path is used for transmitting data that needs to be sent to the first node, the method includes: when the first link fails or recovers from the failure, the network device determines at least one pair of pending nodes from the at least three nodes, Each of the to-be-processed node pairs includes two to-be-processed nodes connected via one link, and the to-be-processed node can send a message to the first node by using a second path that does not include the first link, and The node to be processed can send a message to the first node by using the first path including the first link when the first link is normal, where a second path is from a to-be-processed node to the first node. The path that does not include the total link overhead in the path of the first link is not included, and the first path is from a pending node to the first section when the first link is normal. An optimal path, and each of the to-be-processed nodes in each pair of nodes to be processed has a different up-and-down hop relationship on the second path and the first path; the network device determines a path cost change value of each of the to-be-processed nodes, The path cost change value of each of the to-be-processed nodes is the difference between the first path cost of the to-be-processed node and the second path cost, where the first path cost is the total on the first path when the first link is normal. The link cost, the second path cost is the total link cost on the second path; the network device performs the link cost of the first link at least twice according to the path cost change value of the to-be-processed node. Adjusting, so that when the first link recovers from the failure, the next hop node on the first path of each to-be-processed node is migrated to the optimal path of the first node before the previous hop node Up to the first path, or to make the first hop node on the first path of each to-be-processed node prior to the next hop node to be optimal to the first node when the first link fails Path migration out of the Path all the way.

With reference to the first aspect, in a first implementation manner of the first aspect, the network device performs at least two adjustments on the link cost of the first link according to the path cost change value of the to-be-processed node, including: The network device determines, from the to-be-processed node, the N target nodes, where the number of the target nodes is less than or equal to the number of the to-be-processed nodes; and the network device determines the first chain according to the path cost change value of each of the target nodes. The link cost of the road is adjusted N times.

With reference to the first aspect and the foregoing implementation manner, in a second implementation manner of the first aspect, the network device determines, from the to-be-processed node, the N target nodes, including: the network device, the entire node to be processed As the N target nodes.

In combination with the first aspect and the foregoing implementation manner, in a third implementation manner of the first aspect, The network device adjusts the link cost of the first link N times according to the path cost change value of each target node, including: when the first link recovers from the fault, the network device is in a decreasing manner Performing a first sorting process on the path cost change value of each target node; the network device performs N times adjustment on the link cost of the first link, so that the link of the first link after the i-th adjustment is performed The difference between the cost and the link cost of the first link is smaller than the first value and greater than the second value, where the first value is the i-th path cost change value after the first sorting process, where the The binary value is the largest value among the path cost change values of the first target nodes, the first target node and the second target node form a pair of pending nodes, and the first target node is the second in the first path. a previous hop node of the target node, the second target node is a node corresponding to the i-th path cost change value after the first sorting process, or the second value is the ith after the first sorting process + 1 path overhead change Value.

With reference to the first aspect and the foregoing implementation manner, in a fourth implementation manner of the first aspect, the network device performs N times adjustment on a link cost of the first link according to a path cost change value of each target node. The network device performs a second sorting process on the path cost change value of each target node in an incremental manner when the first link recovers from the fault; the network device is configured to the first link The link cost is adjusted N times, so that the difference between the link cost of the first link and the link cost of the first link is less than the third value and greater than the fourth value, where The third value is the N-th + 1 path cost change value after the second sorting process, and the fourth value is the largest value among the path cost change values of each third target node, the third target The node and the fourth target node form a pair of pending nodes, and the third target node is a previous hop node of the fourth target node in the first path, and the fourth target node is processed by the second sorting process N _ i + 1 The node corresponding to the path cost change value, or the fourth value is the N-th path cost change value after the second sorting process.

With reference to the first aspect and the foregoing implementation manner, in a fifth implementation manner of the first aspect, the network device performs N times adjustment on a link cost of the first link according to a path cost change value of each target node. And the network device performs a third sorting process on the path cost change value of each target node in an incremental manner when the first link fails; the link cost of the network device to the first link Performing N adjustments so that the difference between the link cost of the first link after the i-th adjustment and the link cost of the first link is greater than a fifth value and less than a sixth value, wherein the sixth The value is the i+1th path cost change value after the third sorting process, and the fifth value is the largest value among the path cost change values of the fifth target nodes, and the fifth target node and the sixth target Node Forming a pair of nodes to be processed, and the fifth target node is a previous hop node of the sixth target node in the first path, and the sixth target node is the i+1th path cost after the third sorting process The node corresponding to the change value, or the fifth value is the i-th path cost change value after the third sorting process.

With reference to the first aspect and the foregoing implementation manner, in a sixth implementation manner of the first aspect, the network device performs N times adjustment on a link cost of the first link according to a path cost change value of each target node. And the network device performs a fourth sorting process on the path cost change value of each target node in a decreasing manner on the first link; the link cost of the network device to the first link Performing N adjustments, so that the difference between the link cost of the first link after the i-th adjustment and the link cost of the first link is greater than a seventh value and less than an eighth value, wherein the eighth The value is the N-th path cost change value after the fourth sorting process, and the seventh value is the largest value among the path cost change values of the seventh target nodes, and the seventh target node and the eighth target The node constitutes a pair of nodes to be processed, and the seventh target node is a previous hop node of the eighth target node in the first path, and the eighth target node is the N - i + 1 after the fourth sorting process Path Change value corresponding to the node pin or the seventh value N after the first through the fourth sorting process - i + 1 paths cost change value.

With reference to the first aspect and the foregoing implementation manner, in a seventh implementation manner of the first aspect, the network device performing at least two adjustments on the link cost of the first link includes: determining, by the network device, each target node calculation The processing time required for the optimal path; the network device determines the time interval between the at least two adjustments according to the processing time; the network device is at the time interval, and the link cost of the first link is at least two Adjustments.

In conjunction with the first aspect and the foregoing implementation manner, in an eighth implementation manner of the first aspect, the network device is the second node.

A second aspect provides an apparatus for adjusting link overhead, where the apparatus includes: a to-be-processed node determining unit, configured to determine at least three nodes from a failure when the first link fails or recovers from a failure. a pair of nodes to be processed, wherein each pair of nodes to be processed includes two nodes to be processed connected via one link, and the node to be processed can be to the at least three nodes through a second path that does not include the first link The first node sends a message, and the to-be-processed node can send a message to the first node by using the first path of the first link when the first link is normal, where a second path is a path of the to-be-processed node having the smallest total link overhead in the path to the first node that does not include the first link, and a first path is from the first link when the first link is normal The optimal path of the node to be processed to the first node, and the node to be processed in each pair of nodes to be processed is different from the up and down hop relationship on the second path on the second path, the at least three nodes The second node is directly connected to the first node by using the first link, where the first link is used for transmitting data that needs to be sent to the first node; and the path cost change value determining unit is configured to determine each The path cost change value of the node to be processed, the path cost change value of each node to be processed is the difference between the first path cost of the to-be-processed node and the second path cost, and the first path cost is when the first link is normal. The total link cost on the first path, the second path cost is the total link cost on the second path, and the link cost adjustment unit is configured to change the path cost change value of each to-be-processed node. Performing at least two adjustments on the link cost of the first link, so that when the first link recovers from the failure, the next hop node on the first path of each pair of pending nodes precedes One-hop node The optimal path to the first node is migrated to the first path, or when the first link fails, the previous hop node on the first path of each pending node is prior to the next hop The node migrates the optimal path to the first node out of the first path.

With reference to the second aspect, in a first implementation manner of the second aspect, the link cost adjustment unit is further configured to determine, from the to-be-processed node, N target nodes, where the number of the target nodes is less than or equal to the The number of processing nodes is used to perform N times adjustment on the link cost of the first link according to the path cost change value of each target node.

With reference to the second aspect and the foregoing implementation manner, in the second implementation manner of the second aspect, the link cost adjustment unit is specifically configured to use all of the to-be-processed nodes as the N target nodes.

With reference to the second aspect and the foregoing implementation manner, in a third implementation manner of the second aspect, the link cost adjustment unit is specifically configured to: when the first link recovers from a fault, the network device is decremented And performing a first sorting process on the path cost change value of each target node, and performing N times adjustment on the link cost of the first link, so that the link of the first link after the i-th adjustment is performed The difference between the cost and the link cost of the first link is smaller than the first value and greater than the second value, where the first value is the i-th path cost change value after the first sorting process, where the The binary value is the largest value among the path cost change values of the first target nodes, the first target node and the second target node form a pair of pending nodes, and the first target node is the second in the first path. a previous hop node of the target node, the second target node is a node corresponding to the i-th path cost change value after the first sorting process, or the second value is the ith after the first sorting process + 1 path The cost change value.

In combination with the second aspect and the foregoing implementation manner, in a fourth implementation manner of the second aspect, The link cost adjustment unit is configured to: when the first link recovers from the fault, the network device performs a second sorting process on the path cost change value of each target node in an incremental manner; The link cost of a link is adjusted N times, so that the difference between the link cost of the first link after the first adjustment and the link cost when the first link is normal is less than the third value and greater than the fourth. a value, where the third value is the N-th + 1 path cost change value after the second sorting process, where the fourth value is the largest value of the path cost change values of each third target node, The third target node and the fourth target node form a pair of pending nodes, and the third target node is a previous hop node of the fourth target node in the first path, and the fourth target node is subjected to the second sorting The node corresponding to the processed Nth _i + 1 path cost change value, or the fourth value is the N-th path cost change value after the second sorting process.

With reference to the second aspect and the foregoing implementation manner, in a fifth implementation manner of the second aspect, the link cost adjustment unit is specifically configured to: when the first link fails, the network device is in an incremental manner, The path cost change value of each of the target nodes is subjected to a third sorting process; and the link cost of the first link is adjusted N times, so that the link cost of the first link after the i-th adjustment is When the first link is normal, the difference between the link costs is greater than the fifth value and is less than the sixth value, where the sixth value is the i+1th path cost change value after the third sorting process, the fifth The value is the largest value among the path cost change values of the fifth target nodes, the fifth target node and the sixth target node form a pair of pending nodes, and the fifth target node is the sixth target in the first path a node of the last hop of the node, the sixth target node is a node corresponding to the i +1 path cost change value after the third sorting process, or the fifth value is after the third sorting process i path overhead Change value.

With reference to the second aspect and the foregoing implementation manner, in a sixth implementation manner of the second aspect, the link cost adjustment unit is specifically configured to: when the first link fails, the network device is in a decreasing manner, The path cost change value of each of the target nodes is subjected to a fourth sorting process; and the link cost of the first link is adjusted N times, so that the link cost of the first link after the i-th adjustment is The difference between the link costs of the first link is greater than the seventh value and is less than the eighth value, wherein the eighth value is the N-th path cost change value after the fourth sorting process, and the seventh The value is the largest value among the path cost change values of the seventh target nodes, the seventh target node and the eighth target node form a pair of pending nodes, and the seventh target node is the eighth target in the first path a node of the last hop of the node, the eighth target node is a node corresponding to the N-i + 1 path cost change value after the fourth sorting process, or the seventh value is after the fourth sorting process N-i + 1 path cost change value.

With reference to the second aspect and the foregoing implementation manner, in a seventh implementation manner of the second aspect, the link cost adjustment unit is further configured to determine a processing time required for each target node to calculate an optimal path; Time, determining a time interval between the at least two adjustments; and performing, according to the time interval, performing at least two adjustments on a link cost of the first link.

With reference to the second aspect and the foregoing implementation manner, in an eighth implementation manner of the second aspect, the device is the second node.

A method and apparatus for adjusting link overhead according to an embodiment of the present invention, by adjusting a link cost of a first link directly connected to a first node, including the first link and in the first chain When the path is normal as the first path to the optimal link to the first node, when the first link recovers from the failure, the pair of nodes to be processed that may have micro-ring phenomenon is on the first path. The next hop node migrates the optimal path to the first node to the first path before the previous hop node, or in the pair of pending nodes that may cause the micro ring phenomenon when the first link fails The upper hop node on the first path migrates the optimal path to the first node out of the first path before the next hop node, thereby preventing the micro ring phenomenon from occurring between the pair of nodes to be processed. DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings to be used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only the present invention. For some embodiments, other drawings may be obtained from those of ordinary skill in the art without departing from the drawings.

FIG. 1 is a schematic flowchart of a method for adjusting link overhead according to an embodiment of the present invention.

Fig. 2 is a schematic block diagram showing an example of a communication system to which the method for adjusting link overhead is applied according to an embodiment of the present invention.

3 is a schematic topological view of an inverse optimal path prioritization algorithm when a first node is used as a root in a communication system to which the method for adjusting link overhead is applied according to an embodiment of the present invention.

4 is a schematic block diagram of an apparatus for adjusting link overhead according to an embodiment of the present invention.

FIG. 5 is a schematic structural diagram of an apparatus for adjusting link overhead according to an embodiment of the present invention. detailed description

The technical solution in the embodiment of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention. It is clear that the described embodiments are part of the embodiments of the invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without departing from the inventive scope are the scope of the present invention.

The technical solution of the present invention can be applied to various communication systems, such as: Global System of Mobile Communication (GSM), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access ( WCDMA, Wideband Code Division Multiple Access Wireless), General Packet Radio Service (GPRS), Long Term Evolution (LTE), etc.

FIG. 1 is a schematic flow chart of a method 100 for adjusting link overhead according to an embodiment of the present invention. The method 100 is performed in a communication system including at least three nodes, wherein a first node is directly connected to a second node by using a first link, and the first link is used for transmission to be sent to the first node. Data, as shown in FIG. 1, the method 100 includes:

S110. When the first link fails or recovers from the fault, the network device determines at least one pair of pending nodes from the at least three nodes, where each pair of to-be-processed nodes includes two connected by one link. a node to be processed, the node to be processed can send a message to the first node by using a second path that does not include the first link, and the node to be processed can include the first link when the first link is normal. The first path of a link sends a message to the first node, where a second path is a total link cost from a to-be-processed node to a path of the first node that does not include the first link. a minimum path, a first path is an optimal path from a pending node to the first node when the first link is normal, and each of the pending nodes in the pair of pending nodes is in a second path The upper and lower jumps on the first path are different from each other;

S120, the network device determines a path cost change value of each to-be-processed node, where a path cost change value of each to-be-processed node is a difference between a first path cost and a second path cost of each to-be-processed node, where the first path cost is Is the total link cost on the first path when the first link is normal, and the second path cost is the total link overhead on the second path;

S130, the network device performs at least two adjustments on the link cost of the first link according to the path cost change value of the to-be-processed node, so that each of the pending links is to be processed when the first link recovers from the fault. The next hop node on the first path of the node pair migrates the optimal path to the first node to the first path before the previous hop node, or

When the first link fails, the previous hop node on the first path of each to-be-processed node pair migrates the optimal path to the first node out of the first path before the next hop node . In the prior art, when a link failure or a fault recovery occurs, the link cost of the faulty link is adjusted and released at one time, and each node in the network recalculates the optimal path based on the link cost, because each network device The differences in the hardware capabilities and the differences between the internal and external environments of the device operation result in inconsistencies in the start time, run time, end time of the calculation, and the time point at which the forwarding information table entries are sent, which leads to the microring phenomenon. appear.

In contrast, in the embodiment of the present invention, the link cost of the faulty link is adjusted and released multiple times (at least twice), and when the link is restored, the original optimal path including the faulty link is The one-hop node switches the optimal path back to the original optimal path before the previous node. When the link fails, the upper-hop node on the original optimal path moves the optimal path out of the original one before the next node. path. Therefore, the occurrence of the microring phenomenon can be avoided.

Specifically, in an embodiment of the invention, the method 100 needs to be performed in an IP network comprising a plurality (at least three) of nodes, and Figure 2 shows an example of a communication system 200 to which the method 100 is applicable. As shown in FIG. 2, the communication system 200 includes, for example, 10 nodes, namely: node a, node b, node c, node d, node e, node f, node g, node h, node i, node g.

It should be understood that the configurations of the communication system 200 enumerated above are merely illustrative, and the present invention is not limited thereto, and the number of nodes included and the connection relationship between the nodes may be arbitrarily changed.

For ease of understanding, the node a is used as the first node and the node b is used as the second node, and the link b→a for transmitting the data sent by the node b to the node a is used as the first link. Be explained.

Optionally, in the embodiment of the present invention, the network device is the second node.

Specifically, as the implementation body of the method 100, it may be a second node (for example, node b), or may be used to adjust when the first link (for example, link b→a) fails or recovers from failure. The network device of the link overhead, and the network device can be set independently, or can be set on one or more nodes, and in the embodiment of the present invention, one for each link in the system 200 can be separately configured. The network device for performing the method 100 may also adjust the link cost when the link is faulty or faulty, and the network is not limited.

Hereinafter, for the sake of easy understanding, the node b will be described as the network device without loss of generality.

S110, the node b can detect the working condition of the link b→a, and the link b→a fails. When it is recovered from the fault, it is determined that the method 100 needs to be performed, and the detection method may be the same as or similar to the prior art. Here, in order to avoid redundancy, the description thereof is omitted.

As described above, since the microring phenomenon occurs between the nodes to be processed, that is, the two nodes to be processed have opposite to the upper and lower hops on the selected optimal path before and after the failure of the link b→a, therefore, Node b can determine the nodes (to-be-processed nodes) that need to recalculate and select the optimal path due to the impact of link b→a failure or recovery from failure.

In the system 200 shown in FIG. 2, the link cost of the link b→a is 10, the link cost of the link b→g is 10, and the link cost of the link g→i is 10, the link i→ The link cost of j is 10, the link cost of link j→a is 100, the link cost of link b→c is 10, the link cost of link c→d is 10, and the link d→e The link cost is 10, the link cost of link e→f is 10, the link cost of link f→a is 100, and the link cost of link f→a is 10.

It should be understood that the numerical values of the respective link overheads listed above are merely illustrative, and the present invention is not limited thereto, and the link overhead of each link can be arbitrarily changed.

The path from node b to node a includes: path 1 (ie, link b→a), path 2 (including link b→g, link g→i, link i→j, link j→a), Path 3 (including link b→c, link c→d, link d→e, link e→f, link f→a). The total link cost of the path 1 is the link cost of the link b→a (that is, 10), and the total link cost of the path 2 is the link cost of the link b→g, and the link of the link g→i. The sum of the link cost, the link cost of link i→j, and the link cost of link j→a (ie, 130), the total link cost of path 3 is the link cost and link of link b→c. The link cost of c→d, the link cost of link d→e, the link cost of link e→f, and the link cost of link f→a (ie, 140). Therefore, when link b→a is normal, the optimal path from node b to node a is path 1. When link b→a fails, the optimal path from node b to node a is path 2.

The path from node c to node a includes: path 4 (including link c→b, link b→a), path 5 (including link c→b, link b→g, link g→i, link) i→j, link j→a), path 6 (including link c→d, link d→e, link e→f, link f→a). The total link cost of path 4 is 20, the total link cost of path 5 is 140, and the total link cost of path 6 is 130. Therefore, when link b→a is normal, the optimal path from node c to node a is path 4. When link b→a fails, the optimal path from node c to node a is path 6.

The path from node d to node a includes: path 7 (including link d→c, link c→b, link b→a), path 8 (including link d→c, link c→b, link) b→g, link g→i, link i →j, link j→a), path 9 (including link d→e, link e→f, link f→a). Through the summation calculation as described above, it can be determined that the total link cost of the path 7 is 30, the total link cost of the path 8 is 150, and the total link cost of the path 9 is 120. Therefore, when link b→a is normal, the optimal path from node d to node a is path 7. When link b→a fails, the optimal path from node d to node a is path 9.

The path from node e to node a includes: path 10 (including link 6 → (1, link d→c, link c→b, link b→a), path 11 (including link e→d, chain) Road d→c, link c→b, link b→g, link g→i, link i→j, link j→a), path 12 (including link e→f, link f→ a) wherein, by the summation calculation as described above, it can be determined that the total link cost of the path 10 is 40, the total link cost of the path 11 is 160, and the total link cost of the path 12 is 110. Therefore, when the chain When the path b→a is normal, the optimal path from node e to node a is path 10. When link b→a fails, the optimal path from node e to node a is path 12.

The path from node f to node a includes: path 13 (including link f→e, link e→d, link d→c, link c→b, link b→a), path 14 (including link) f→e, link e→d, link d→c, link c→b, link b→g, link g→i, link i→j, link j→a), path 15 ( Includes link f→a). Through the summation calculation as described above, it can be determined that the total link cost of the path 13 is 50, the total link cost of the path 14 is 170, and the total link cost of the path 15 is 100. Therefore, when link b→a is normal, the optimal path from node f to node a is path 13. When link b → & fails, the optimal path from node f to node a is path 15.

The path from node g to node a includes: path 16 (including link g→b, link b→a), path 17 (including link g→i, link i→j, link j→a), path 18 (including link g→b, link b→c, link c→d, link d→e, link e→f, link f→a). The total link cost of the path 16 is 20, the total link cost of the path 17 is 120, and the total link cost of the path 18 is 150 by the summation calculation as described above. Therefore, when link b→a is normal, the optimal path from node g to node a is path 16. When link b→a fails, the best path from node g to node a is path 17.

The path from node h to node a includes: path 19 (including link h→g, link g→b, chain^b→a), path 20 (including link h→g, link g→i, link) i→j, link j→a), path 21 (including link h→g, link g→b, link b→c, link c→d, link d→e, link e→f , link f→a). Through the summation calculation as described above, it can be determined that the total link cost of the path 19 is 30, the total link cost of the path 20 is 130, and the total link cost of the path 21 is 160. Therefore, when link b→a is normal, the optimal path from node h to node a is path 19. When link b→a fails, the optimal path from node h to node a is path 20.

The path from node i to node a includes: path 22 (including link i→g, link g→b, link b→a), path 23 (including link i→j, link j→a), path 24 (including link i→g, link g→b, link b→c, link c→d, link d→e, link e→f, link f→a). The total link cost of the path 22 is 30, the total link cost of the path 23 is 110, and the total link cost of the path 24 is 160 by the summation calculation as described above. Therefore, when link b→a is normal, the optimal path from node i to node a is path 22. When link b→a fails, the optimal path from node i to node a is path 23.

The path from node j to node a includes: path 25 (including link j→i, link i→g, link g→b, link b→a), path 26 (including chain j→a), path 27 (including link j→i, link i→g, link g→b, link b→c, link c→d, link d→e, link e→f, link f →a) . The total link cost of the path 25 is 40, the total link cost of the path 26 is 100, and the total link cost of the path 27 is 170 by the summation calculation as described above. Therefore, when link b→a is normal, the optimal path from node j to node a is path 25. When link b→a fails, the optimal path from node j to node a is path 26.

As described above, for the above-mentioned node b, node c, node d, node e, node f, node g, node h, node i, node j, when link b→a fails or recovers from failure, to node a The optimal path changes, and when the link b→a is normal, the optimal link includes the link b→a. Therefore, it can be determined that in the system 200, the node b, the node c, the node d, the node e, the node f, the node g, the node h, the node i, and the node j are subjected to a failure or failure recovery of the link b→a. The affected node (the node to be processed).

It should be understood that since the network architecture changes with the connection between nodes and the increase or decrease of nodes, the nodes affected by the first link also change when different architectures or architectures change. Therefore, the above-mentioned pending nodes are listed. For illustrative purposes only, the present invention is not limited thereto, and the affected node may be selected according to a network architecture.

After determining the affected node, the pair of pending nodes that may have a micro-ring phenomenon may be determined from the above-mentioned affected nodes, for example, in path 4 (the first path of node c), node b is node c Next hop node, in path 2 (the second path of node b), node b is the last hop node of node c, that is, before and after the link b→a failure, the bounce relationship between node b and node c occurs. Change, micro-ring phenomenon may occur between node b and node c, therefore, node b and node C constitutes a pair of processing nodes.

Similarly, node c and node b constitute a pair of nodes to be processed; node d and node c constitute a pair of nodes to be processed, node e and node d constitute a pair of nodes to be processed, node f and node e constitute a pair For the pair of processing nodes, node b and node g constitute a pair of processing nodes, node g and node i form a pair of nodes to be processed, and node i and node j constitute a pair of nodes to be processed.

It should be noted that, in the embodiment of the present invention, in the path 19 (the first path of the node h) and the path 20 (the second path of the node h), the next hop node of the node h is the node g, so the node There is no microring problem between h and node g. Therefore, node h is not treated as a node to be processed.

It should be understood that, in the embodiment of the present invention, the "first path" refers to a node to one end node of the faulty link (for example, node a) when the faulty link (for example, link b→a) is normal. The optimal path, and the path takes link b→a as a segment of the path. Therefore, for different nodes, the first path is different.

The "second path" refers to an optimal path from one node to one end node (for example, node a) of the faulty link when the faulty link (for example, link b→a) fails, due to the link b→ a A fault occurs at this time, so that the path does not take link b→a as a segment of the path. Therefore, for different nodes, the second path is different.

It should be noted that, in the embodiment of the present invention, a pair of nodes to be processed are connected by a link between two nodes to be processed, and the node to be processed determined as described above (or, in the first The first node (here, node a) is not included in the node affected by the link failure or failure recovery.

S120, the node b may use the node in the pair of nodes to be processed as the node to be processed, and the node b may determine a handover threshold (path cost change value) of the to-be-processed node, where the "switching threshold" refers to triggering path switching. The critical value, that is, when the link cost of link b→a is greater than the handover threshold, the affected node does not select the path including link b→a as the optimal path to node a, when link b When the link cost of →a is less than the handover threshold, the affected node selects the path including link b→a as the optimal path to node a.

In the embodiment of the present invention, the total link cost (second path cost) of the to-be-processed node on the optimal path when the first link fails, and the optimal path when the first link is normal may be used. The difference between the total link overhead (first path overhead) on the switch threshold of the to-be-processed node. Table 1 below shows each node to be processed in system 200 (node b, node c, node d, node e, Switching threshold for node f, node g, node i, node j).

Table 1

At S130, the node b may perform multiple (at least two) adjustments on the link cost of the link b→a according to the threshold value corresponding to each to-be-processed node determined as described above, so that after each adjustment, For the previous hop node and the next hop node on the same path, the optimal path is not switched at the same time.

In the embodiment of the present invention, when the link b→a recovers from the failure (ie, case 1) and when the link b→a fails (ie, case 2), the link to the link b→a The adjustment method of the overhead is different. The following describes the adjustment actions and methods in the above two cases.

Situation 1

When the link b→a recovers from the failure, the node b can adjust the link overhead of the link b→a from the preset maximum value (for example, 16777215) to the normal value (here, 10).

Optionally, in the embodiment of the present invention, the network device performs at least two adjustments on the link cost of the first link according to the path cost change value of the to-be-processed node, including:

The network device determines N target nodes from the to-be-processed node;

The network device adjusts the link cost of the first link N times according to the path cost change value of each target node.

In the embodiment of the present invention, the number of adjustments (the same as the number N of target nodes) may be determined according to the number of nodes to be processed. Optionally, the determining, by the network device, the at least one target node from the to-be-processed node, the method includes: the network device as the N target nodes.

Specifically, in the embodiment of the present invention, the number of adjustments may be the same as the number of nodes to be processed. In the system 200, the number of nodes to be processed is 8, and therefore, for example, 8 times, for example, optionally, the network device changes the value of the path cost of each target node to the chain of the first link. The road cost is adjusted N times, including:

When the first link recovers from the fault, the network device performs a first sorting process on the path cost change value of each target node in a decreasing manner;

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is less than or equal to The first value is greater than the second value, where the first value is the i-th path cost change value after the first sorting process, and the second value is the i-th 1 after the first sorting process Path cost change value.

And optionally, the network device determines, from the to-be-processed node, the at least one target node, including:

The network device determines N target nodes from the to-be-processed nodes, and the path cost change values of the N target nodes are different from each other.

Specifically, in the foregoing embodiment, the switching thresholds of the nodes to be processed are different, but the switching thresholds of the nodes to be processed are not duplicated. When the switching thresholds of the nodes to be processed are repeated, only A pending node whose switching threshold has been repeated is reserved, or only one of the repeated switching thresholds may be reserved.

Node b can sort the switching thresholds of the target node in descending order, that is, the following sorting order can be obtained:

120 (switching threshold for node b), 110 (switching threshold for node c), 100 (switching threshold for node g and node h), 90 (switching threshold for node d), 80 (switching threshold for node i), 70 ( Switching threshold for node e), 60 (switching threshold for node j), 50 (switching threshold for node f).

Thereafter, node b can adjust the link overhead of link b→a 8 times (the number of times is the same as the number of handover thresholds that do not overlap each other in Table 1).

The link cost of the link b→a that is first adjusted and released may be any value less than 130 (120 + 10) and greater than 120 (110 + 10), for example, 125 (115 + 10), after the first time After adjustment, only node b will switch the optimal path to node a to path 2 (including link b→a), The link cost change (115) of the link b→a after the first adjustment is greater than the switching threshold of the node c and the node g, therefore, the node c does not switch the optimal path to the node a to the path 6, Moreover, the node g does not switch the optimal path to the node a to the path 17, thereby avoiding the relationship between the node c and the node b or the node g due to the hardware capability of the node g or the node c and the processing environment being better than the node b. A microring phenomenon occurs between the node b and the node b.

The link cost of the link b→a adjusted and released for the second time may be any value less than 120 (110 + 10) and greater than 110 (100 + 10), for example, 115 (105 + 10), after the second time After adjustment, node c will switch the optimal path to node a to path 6 (including link b→a), because the link cost change (105) of the link b→a after the second adjustment is greater than the node. The switching threshold of d, therefore, node d does not switch the optimal path to node a to path 9, thereby avoiding the relationship between node d and node c due to the hardware capability of node d and the processing environment being better than node c. A microring phenomenon occurs. Moreover, since the node b has switched the optimal path to the node a to the path 2 after the first adjustment, the microring phenomenon does not occur between the node b and the node c.

The link cost of the link b→a adjusted and released for the third time may be any value less than 110 (100 + 10) and greater than 100 (90 + 10), for example, 105 (95 + 10), after the third time After adjustment, node g will switch the optimal path to node a to path 17 (including link b→a), since the link cost change (95) of the link b→a after the third adjustment is greater than the node. The switching threshold of i, therefore, node i does not switch the optimal path to node a to path 23, thereby avoiding the relationship between node i and node g due to the hardware capability of node i and the processing environment being better than node g A microring phenomenon occurs. Also, since the node b has switched the optimal path to the node a to the path 2 after the first adjustment, the microring phenomenon does not occur between the node b and the node g.

It should be noted that, in the embodiment of the present invention, the handover threshold of the node h and the handover threshold of the node g are both 100. Therefore, the node h and the node g simultaneously switch the optimal link to the node a. However, since the handover threshold of the node h is the same as the handover threshold of the node g, it is indicated that in the path 19 and the path 20, the node h is the previous hop node of the node g, so there is no micro-ring problem between the node h and the node g. .

The link cost of the link b→a adjusted and released for the fourth time may be any value less than 100 (90 + 10) and greater than 90 (80 + 10), for example, 95 (85 + 10), after the fourth time After adjustment, node d will switch the optimal path to node a to path 9 (including link b→a), because the link cost change (85) of link b→a after the fourth adjustment is greater than the node. The switching threshold of e, therefore, the node e does not switch the optimal path to the node a to the path 12, thereby avoiding the node e The hardware capability and processing environment are better than the node d, which causes a microring phenomenon between the node e and the node d. And, since the node c has switched the optimal path to the node a to the path 6 after the second adjustment, the micro-ring phenomenon does not occur between the node c and the node d.

The link cost of link b→a, which is adjusted and released for the fifth time, can be any value less than 90 (80 + 10) and greater than 80 (70+ 10), for example, 85 (75 + 10), after the fifth time After adjustment, node i will switch the optimal path to node a to path 23 (including link b→a), because the link cost change (75) of link b→a after the fifth adjustment is greater than the node. The switching threshold of j, therefore, node j does not switch the optimal path to node a to path 26, thereby avoiding the relationship between node j and node i due to the hardware capability of node j and the processing environment being better than node i A microring phenomenon occurs. And, since the node g has switched the optimal path to the node a to the path 17 after the third adjustment, the micro-ring phenomenon does not occur between the node g and the node i.

The link cost of link b→a, which is adjusted and released for the sixth time, can be any value less than 80 (70 + 10) and greater than 70 (60+ 10), for example, 75 (65 + 10), after the sixth time. After adjustment, node e will switch the optimal path to node a to path 12 (including link 1) → 3), because the link cost change (65) of link b→a after the sixth adjustment is greater than The switching threshold of the node f, therefore, the node f does not switch the optimal path to the node a to the path 15, thereby avoiding the node f and the node e due to the hardware capability of the node f and the processing environment being better than the node e A microring phenomenon occurs between them. Moreover, since the node d has switched the optimal path to the node a to the path 9 after the fourth adjustment, the micro-ring phenomenon does not occur between the node d and the node e.

The link cost of the link b→a adjusted and released for the seventh time may be any value less than 70 (60+10) and greater than 60 (50 + 10), for example, 65 (55 + 10), after the seventh time After adjustment, node j will switch the optimal path to node a to path 26 (including link b→a), since node i has switched the optimal path to node a to path 23 after the fifth adjustment, Therefore, the micro-ring phenomenon does not occur between the node i and the node j.

The link cost of link b→a, which is adjusted and released for the eighth time, can be any value less than 60 (50 + 10), for example, 55 (45 + 10). After the eighth adjustment, node f will The optimal path of node a is switched to path 15 (including link 1) → 3). Since node e has switched the optimal path to node a to path 12 after the sixth adjustment, node e and node f For example, the network device adjusts the link cost of a link N times according to the path cost change value of each target node, including: When the first link recovers from the fault, the network device performs a second sorting process on the path cost change value of each target node in an incremental manner;

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is smaller than the first The third value is greater than the fourth value, wherein the third value is the Nth_1 + 1 path cost change value after the second sorting process, and the fourth value is the Nth after the second sorting process - i path cost change values.

Node b can sort the switching thresholds of the target node in increasing order, that is, the following sorting order can be obtained:

50 (switching threshold of node f), 60 (switching threshold of node j), 70 (switching threshold of node e), 80 (switching threshold of node i), 90 (switching threshold of node d), 100 (node g and Switching threshold for node h), 110 (switching threshold for node c), 120 (switching threshold for node b).

The link cost of the link b→a that is first adjusted and released may be any value less than 130 (120 + 10) and greater than 120 (110 + 10), for example, 125 (115 + 10), after the first time After adjustment, only node b will switch the optimal path to node a to path 2 (including link b→a), because the link cost change (115) of link b→a after the first adjustment is greater than The switching threshold of node c and node g, therefore, node c does not switch the optimal path to node a to path 6, and node g does not switch the optimal path to node a to path 17, thereby enabling Avoid micro-ring phenomenon between node c and node b or between node g and node b because the hardware capability and processing environment of node g or node c are better than node b.

The link cost of link b→a that is adjusted and released for the second time can be less than 120 (110 + 10) And any value greater than 110 (100 + 10), for example, 115 (105 + 10), after the second adjustment, node c will switch the optimal path to node a to path 6 (including link b→a) ), since the link cost change (105) of the link b→a after the second adjustment is greater than the switching threshold of the node d, the node d does not switch the optimal path to the node a to the path 9, thereby The micro-ring phenomenon between the node d and the node c can be avoided due to the hardware capability of the node d and the processing environment being better than the node c. Moreover, since the node b has switched the optimal path to the node a to the path 2 after the first adjustment, the micro-ring phenomenon does not occur between the node b and the node c.

The link cost of the link b→a that is adjusted and released for the third time can be any value less than 110 (100+ 10) and greater than 100 (90+ 10), for example, 105 (95 + 10), after the third time After adjustment, node g will switch the optimal path to node a to path 17 (including link b→a), since the link cost change (95) of the link b→a after the third adjustment is greater than the node. The switching threshold of i, therefore, node i does not switch the optimal path to node a to path 23, thereby avoiding the relationship between node i and node g due to the hardware capability of node i and the processing environment being better than node g A microring phenomenon occurs. Also, since the node b has switched the optimal path to the node a to the path 2 after the first adjustment, the microring phenomenon does not occur between the node b and the node g.

The link cost of link b→a, which is adjusted and released for the fourth time, can be any value less than 100 (90 + 10) and greater than 90 (80+ 10), for example, 95 (85 + 10), after the fourth time After adjustment, node d will switch the optimal path to node a to path 9 (including link b→a), because the link cost change (85) of link b→a after the fourth adjustment is greater than the node. The switching threshold of e, therefore, node e does not switch the optimal path to node a to path 12, thereby avoiding the relationship between node e and node d due to the hardware capability of node e and the processing environment being better than node d A microring phenomenon occurs. And, since the node c has switched the optimal path to the node a to the path 6 after the second adjustment, the micro-ring phenomenon does not occur between the node c and the node d.

The link cost of link b→a, which is adjusted and released for the fifth time, can be any value less than 90 (80+ 10) and greater than 80 (70+ 10), for example, 85 (75 + 10), after the fifth time After adjustment, node i will switch the optimal path to node a to path 23 (including link b→a), due to the fifth The link cost change (75) of the second adjusted link b→a is greater than the switching threshold of the node j. Therefore, the node j does not switch the optimal path to the node a to the path 26, thereby avoiding the node j The hardware capability and processing environment are better than node i, which causes micro-ring phenomenon between node j and node i. And, since the node g has switched the optimal path to the node a to the path 17 after the third adjustment, the micro-ring phenomenon does not occur between the node g and the node i.

The link cost of link b→a, which is adjusted and released for the sixth time, can be any value less than 80 (70 + 10) and greater than 70 (60+ 10), for example, 75 (65 + 10), after the sixth time. After adjustment, node e will switch the optimal path to node a to path 12 (including link 1) → 3), because the link cost change ( 65 ) of link b→a after the sixth adjustment is greater than The switching threshold of the node f, therefore, the node f does not switch the optimal path to the node a to the path 15, thereby avoiding the node f and the node e due to the hardware capability of the node f and the processing environment being better than the node e A microring phenomenon occurs between them. Moreover, since the node d has switched the optimal path to the node a to the path 9 after the fourth adjustment, the micro-ring phenomenon does not occur between the node d and the node e.

The link cost of link b→a, which is adjusted and released for the eighth time, can be any value less than 60 (50 + 10), for example, 55 (45 + 10). After the eighth adjustment, node f will The optimal path of node a is switched to path 15 (including link 1) → 3). Since node e has switched the optimal path to node a to path 12 after the sixth adjustment, node e and node f The embodiment of determining the number of adjustments according to the node to be processed is listed, but the present invention is not limited thereto. For example, when the link b→a recovers from the failure, the adjustment may be performed once. And causing the next hop node on the first path of the two or more pairs of pending nodes to switch the optimal path to the first path before the previous hop node. As an implementation method, the following actions can be performed.

That is, the network device optionally adjusts the link cost of the first link at least twice according to the path cost change value of the to-be-processed node, including:

The network device determines the adjustment range of each node to be processed according to the path cost change value of each node to be processed, where the adjustment range of a node to be processed is less than or equal to the path of the node to be processed. The path cost change value is greater than or equal to the path cost change value of the reference node of the to-be-processed node, and the reference node of the to-be-processed node is the path cost change value of the node to be processed in the previous hop node on each first path. Node

The network device adjusts the link overhead of the first link at least twice according to the adjustment range of the to-be-processed node.

Specifically, FIG. 3 is a schematic topological diagram of a reverse optimal path prioritization algorithm when the first node is used as a root in the communication system to which the method for adjusting link overhead is applied according to an embodiment of the present invention. . It should be noted that the digital identifier in FIG. 3 identifies the handover threshold of the node to be processed. When the link b→a recovers from the fault, it is necessary to ensure that the next hop node in each pair of pending nodes (the hop count from the node a is smaller) precedes the previous hop node (the number of hops from the node a is higher. Large) switching the optimal path, and since the first path of each node includes the link b→a, or the first path uses the node b as the last transit node, it is necessary to ensure the distance from the node a. The node with the smallest number of hops (ie, node b) completes the handover first.

In the embodiment of the present invention, first, in the case of the first link failure, the reverse shortest path first (RSPF) algorithm is used to calculate the path cost of each node to the root node (first node). The value D(i,max). For the link normal, use the RSPF algorithm to calculate the cost value D ( i, min ) from each node to the root node.

Thereafter, the maximum link cost value Δ cost ( i , max ) of the first link that needs to be adjusted for the anti-micro ring switching of each node to be processed is calculated:

A cost ( i, max ) = D(i, max) - D ( i, min ).

Calculate the minimum link cost Δ cost ( i, min ) of the first link that needs to be adjusted for the anti-micro ring switching of each node to be processed:

Δ cost ( i, min ) =MAX{ Δ cost ( j , max ) } , where node j is the son node of node i under normal conditions of the first link (or node j is node i in the first path) The last hop node).

Therefore, the range of cost values (ie, the adjustment range) of the first link that needs to be adjusted for the anti-micro ring switching of the node i is <厶. (^ ( i, min ) +K, A cost ( i, max ) + K>, where K is the cost value when the first link is normal.

Thereafter, adjustment can be made according to the adjustment range of each node. For example, in system 200, the adjustment range of node g is [80, 100], the adjustment range of node c is [90, 110], and node g constitutes The switching threshold of the node of the pair of nodes to be processed is not between the switching thresholds of node c and node d. (That is, the switching threshold of the node b is greater than 110, and the switching threshold of the node i is 80 is less than 90). When the link overhead is adjusted to enable the node c to switch the optimal path, the node g can be switched at the same time.

As shown in FIG. 3, since the paths of the remaining nodes to the node b may be different, for example, the nodes (, the node d, the node e, and the node f need to be sequentially switched in the order of the hops from the node a, and The node g, the node i, and the node j also need to be switched in order from the hop count of the node a. Therefore, for the nodes respectively located on the two paths, there may be an initial adjustment to the most The situation of the excellent path.

As described above, in system 200, the handover threshold (i.e., 100) of node g falls between a handover threshold of a pair of nodes to be processed (i.e., node c and node d) (i.e., [90, 110]). And, the switching threshold of the node g constituting the node pair to be processed is not between the switching thresholds of the node c and the node d (that is, the switching threshold of the node b is greater than 110, and the switching threshold of the node i is 80 less than 90). Then, when the link overhead is adjusted so that the node c switches the optimal path, the node g can be switched at the same time.

Similarly, node i and node d can switch the optimal path at the same time, and node j and node e can switch the optimal path at the same time.

That is, node g, node i, and node j can be deleted from the node to be processed as a non-target node.

Without loss of generality, the following methods can be used to determine non-target nodes from the nodes to be processed.

1. Initialize three lists: TentList, CandList, and OutPutList.

2. Link failure The nearest node (here, node b) is the first to do anti-micro ring switching, i has the node self.

3. Push the node self into the TentList.

4. Obtain a node x from the TentList, where node x is the node with the largest lower limit of the adjustment range in the TentList (or the link cost change value of the previous hop node on the first path). If the TentList is empty, the algorithm ends.

5. Push node X into OutPutList.

6. Add all the child nodes y (here, node c and node g) of node X that may have micro-ring switching to CandList.

7. Determine the maximum switching threshold cost 1 of the switching thresholds of all nodes in the CandList, With the switching threshold cost 2 of node z in the TentList, if cost 2> cost 1 , the node z is deleted from the TentList.

8. If the node z is deleted from the TentList, then all the nodes z of the node z that may have a micro-ring switch (which forms a pair of nodes to be processed with the node z) are added to the CandList.

9. Repeat 7~8 until the switching threshold of all nodes in TentList is compared with cost 1 and deleted.

10. Move all the nodes in the CandList to the TentList, making the CandList empty.

11. Jump execution 4.

Thus, it can be determined that each node in the OutPutList is the target node.

Here, in the system 200, each node in the OutPutList determined by the above method is a node b, a node c, a node d, a node e, and a node f.

Then, the switching thresholds of the nodes in the OutPutList may be sorted (incremented or decremented), and the link cost of the link b→a is adjusted and released according to the switching thresholds, and the specific process may be processed according to the foregoing The process of adjusting and releasing the link cost of the link b→a is similar to the switching threshold of the node. Here, in order to avoid redundancy, the description thereof is omitted.

In addition, when the link cost of the link b→a is adjusted and released according to the switching threshold of each node in the OutPutList, the following adjustment method can also be adopted:

That is, the network device adjusts the link cost of the first link N times according to the path cost change value of each target node, including:

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is smaller than the first a value that is greater than the second value, wherein the first value is an i-th path cost change value after the first sorting process,

The second value is the largest value among the path cost change values of the first target nodes, and the first target node and the second target node form a pair of pending nodes, and the first target node is in the first path. a previous hop node of the second target node, where the second target node is a node corresponding to the i-th path cost change value after the first sorting process.

Specifically, after sorting the switching thresholds of the nodes in the OutPutList (here, in descending order as an example), T adjustments can be made, and T is the section included in the OutPutList. The number of points, as described above, in system 200, T is five.

That is, the link cost of the link b→a that is first adjusted and released may be any value less than 130 (120 + 10) and greater than 120 (110 + 10), for example, 121 (111 + 10), by the first After an adjustment, only node b will switch the optimal path to node a to path 2 (including link b→a), due to the change in link cost of link b→a after the first adjustment ( 111 ) is greater than the switching threshold of node c and node g, therefore, node c does not switch the optimal path to node a to path 6, and node g does not switch the optimal path to node a to path 17, thus It can be avoided that the micro-ring phenomenon occurs between node c and node b or between node _g and node b because the hardware capability and processing environment of node g or node c are better than node b.

The link cost of the link b→a adjusted and released for the second time may be any value less than 120 (110 + 10) and greater than 100 (90 + 10), for example, 101 (91 + 10), after the second time After adjustment, node c will switch the optimal path to node a to path 6 (including link b→a), and node g will switch the optimal path to node a to path 17 (including link b→ a). Since the link cost change (91) of the link b→a after the second adjustment is greater than the handover threshold of the node d, the node d does not switch the optimal path to the node a to the path 9, thereby enabling Avoid micro-ring phenomenon between node d and node c because the hardware capability and processing environment of node d are better than node c. Moreover, since the node b has switched the optimal path to the node a to the path 2 after the first adjustment, the microring phenomenon does not occur between the node b and the node c. Moreover, since the link cost change (91) of the link b→a after the second adjustment is greater than the handover threshold of the node i, the node i does not switch the optimal path to the node a to the path 23, thereby The micro-ring phenomenon between node i and node g can be avoided due to the hardware capability of the node i and the processing environment being better than the node g. Also, since the node b has switched the optimal path to the node a to the path 2 after the first adjustment, the microring phenomenon does not occur between the node b and the node g.

It should be noted that, as shown in FIG. 3, in the system 200, the node c has only one child node, but the present invention is not limited thereto. When the node c has a plurality of child nodes, the adjusted The link cost of link b→a is greater than the largest value of the path cost change value of each child node of node c.

It should be noted that, in the embodiment of the present invention, the handover threshold of the node h and the handover threshold of the node g are both 100. Therefore, the node h and the node g simultaneously switch the optimal link to the node a. However, since the handover threshold of the node h is the same as the handover threshold of the node g, it is indicated that in the path 19 and the path 20, the node h is the previous hop node of the node g, so the node h and the node g are not There is a microring problem.

The link cost of the link b→a adjusted and released for the third time may be any value less than 100 (90 + 10) and greater than 80 (70 + 10), for example, 81 (71 + 10), after the third time After adjustment, node d will switch the optimal path to node a to path 9 (including link b→a), because the link cost change (71) of the link b→a after the third adjustment is greater than the node. The switching threshold of e, therefore, node e does not switch the optimal path to node a to path 12, thereby avoiding the relationship between node e and node d due to the hardware capability of node e and the processing environment being better than node d A microring phenomenon occurs. And, since the node c has switched the optimal path to the node a to the path 6 after the second adjustment, the micro-ring phenomenon does not occur between the node c and the node d. And, after the third adjustment, node i will switch the optimal path to node a to path 23 (including link b→a), due to the link cost of link b→a after the third adjustment. The change (71) is greater than the switching threshold of the node j. Therefore, the node j does not switch the optimal path to the node a to the path 26, thereby avoiding the node due to the hardware capability of the node j and the processing environment being better than the node i. A microring phenomenon occurs between j and node i. And, since the node g has switched the optimal path to the node a to the path 17 after the second adjustment, the micro-ring phenomenon does not occur between the node g and the node i.

It should be noted that, as shown in FIG. 3, in the system 200, the node d has only one child node, but the present invention is not limited thereto. When the node d has a plurality of child nodes, the adjusted The link cost of link b→a is greater than the largest value of the path cost change value of each child node of node d.

The link cost of the link b→a adjusted and released for the fourth time may be any value less than 80 (70 + 10) and greater than 60 (50 + 10), for example, 61 ( 51 + 10 ), after the fourth time After adjustment, node e will switch the optimal path to node a to path 12 (including link 1) →& ), since the link cost change ( 61 ) of the link b→a after the fourth adjustment is greater than The switching threshold of the node f, therefore, the node f does not switch the optimal path to the node a to the path 15, thereby avoiding the node f and the node e due to the hardware capability of the node f and the processing environment being better than the node e. A microring phenomenon occurs between them. Moreover, since the node d has switched the optimal path to the node a to the path 9 after the third adjustment, the micro-ring phenomenon does not occur between the node d and the node e. And, after the fourth adjustment, node j will switch the optimal path to node a to path 26 (including link 1) → 3), since node i has reached the most of node a after the third adjustment The optimal path is switched to path 23, so the micro-ring phenomenon does not occur between node i and node j.

It should be noted that, as shown in FIG. 3, in the system 200, the node e has only one child node. However, the present invention is not limited thereto. When the node e has multiple child nodes, the link cost of the adjusted link b→a may be greater than the largest path cost change value of each child node of the node e. value.

The link cost of link b→a, which is adjusted and released for the fifth time, can be any value less than 60 ( 50 + 10 ), for example, 55 ( 45 + 10 ). After the fifth adjustment, node f will The optimal path of node a is switched to path 15 (including link 1) → 3). Since node e has switched the optimal path to node a to path 12 after the fourth adjustment, node e and node f The micro-rings do not occur in the same way. The switching thresholds of the nodes in the OutPutList can be arranged in an ascending order and adjusted. The process is similar to the above process. Here, in order to avoid redundancy, the description is omitted.

When the first link recovers from the failure, the network device performs a second sorting process on the path cost change value of each target node in an incremental manner;

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is smaller than the first The third value is greater than the fourth value, wherein the third value is the Nth_1 + 1 path cost change value after the second sorting process.

The fourth value is the largest value among the path cost change values of the third target nodes, and the third target node and the fourth target node form a pair of pending nodes, and the third target node is in the first path. a previous hop node of the fourth target node, where the fourth target node is a node corresponding to the Nth_1 + 1 path cost change value after the second sorting process.

According to the method for adjusting the link overhead according to the embodiment of the present invention, by performing the deletion processing on the node to be processed, the number of times of processing can be reduced, and the efficiency of the adjustment can be improved, thereby improving the practicability of the adjustment.

Situation 2

When the link b→a fails, the node b can adjust the link overhead of the link b→a from the normal value (here, 10) to the preset highest value (for example, 16777215).

The network device determines N target nodes from the to-be-processed node; The network device adjusts the link cost of the first link N times according to the path cost change value of each target node.

In the embodiment of the present invention, the number of adjustments (the same as the number N of target nodes) may be determined according to the number of nodes to be processed.

Optionally, the network device determines, from the to-be-processed node, the at least one target node, where: the network device uses all of the to-be-processed nodes as the N target nodes.

When the first link fails, the network device performs a fourth sorting process on the path cost change value of each target node in a decreasing manner;

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is greater than The seventh value is less than the eighth value, wherein the seventh value is the N - i + 1 path cost change value after the fourth sorting process, and the eighth value is the Nth after the fourth sorting process _ i path cost change values.

120 (switching threshold for node b), 110 (switching threshold for node c), 100 (switching threshold for node g and node h), 90 (switching threshold for node d), 80 (switching threshold for node i), 70 ( Switching threshold for node e), 60 (switching threshold for node j), 50 (switching threshold for node f). Thereafter, the node b can adjust the link overhead of the link b→a 8 times (the number of times is the same as the number of handover thresholds not repeated in Table 1).

The link cost of the link b→a that is adjusted and released for the first time can be any value less than 70 (60 + 10) and greater than 60 (50+ 10), for example, 65 (55 + 10), after the first time After adjustment, only node f will switch the optimal path to node a to path 13 (excluding link b→a), due to the change in link cost of link b→a after the first adjustment (55) It is smaller than the switching threshold of the node e. Therefore, the node e does not switch the optimal path to the node a to the path 10, thereby avoiding the node e and the node f due to the hardware capability of the node e and the processing environment being better than the node f. A microring phenomenon occurs between them.

The link cost of the second adjusted and released link b→a can be any value less than 80 (70 + 10) and greater than 70 (60+ 10), for example, 75 (65 + 10), after the second time After adjustment, node j will switch the optimal path to node a to path 25 (excluding link b→a), because the link cost change (65) of link b→a after the second adjustment is less than The switching threshold of the node i, therefore, the node i does not switch the optimal path to the node a to the path 110, thereby avoiding the node i and the node j due to the hardware capability of the node i and the processing environment being better than the node j A microring phenomenon occurs between them.

The link cost of link b→a, which is adjusted and released for the third time, can be any value less than 90 (80 + 10) and greater than 80 (70+ 10), for example, 85 (75 + 10), after the third time After adjustment, node e will switch the optimal path to node a to path 10 (excluding link b→a), because the link cost change (75) of link b→a after the third adjustment is less than The switching threshold of the node d, therefore, the node d does not switch the optimal path to the node a to the path 7, thereby avoiding the node d and the node e due to the hardware capability of the node d and the processing environment being better than the node e A microring phenomenon occurs between them. Moreover, since the node f has switched the optimal path to the node a to the path 13 after the first adjustment, the micro-ring phenomenon does not occur between the node e and the node f.

The link cost of link b→a, which is adjusted and released for the fourth time, can be any value less than 100 (90 + 10) and greater than 90 (80+ 10), for example, 95 (85 + 10), after the fourth time After adjustment, node i will switch the optimal path to node a to path 22 (excluding link b→a), because the link cost change (85) of link b→a after the fourth adjustment is less than The switching threshold of the node g, therefore, the node g does not switch the optimal path to the node a to the path 16, thereby avoiding the node g and the node i due to the hardware capability of the node g and the processing environment being better than the node i A microring phenomenon occurs between them. And, since the node j has switched the optimal path to the node a to the path 25 after the second adjustment, the micro-ring phenomenon does not occur between the node i and the node j. The link cost of link b→a, which is adjusted and released for the fifth time, can be any value less than 110 (100+ 10) and greater than 100 (90+ 10), for example, 105 (95 + 10), after the fifth time After adjustment, node d will switch the optimal path to node a to path 7 (excluding link b→a), because the link cost change (95) of link b→a after the fifth adjustment is less than The switching threshold of the node c, therefore, the node c does not switch the optimal path to the node a to the path 4, thereby avoiding the node c and the node d due to the hardware capability of the node c and the processing environment being better than the node d A microring phenomenon occurs between them. Moreover, since the node e has switched the optimal path to the node a to the path 10 after the third adjustment, the micro-ring phenomenon does not occur between the node d and the node e.

The link cost of link b→a, which is adjusted and released for the sixth time, can be any value less than 120 (110+10) and greater than 110 (100 + 10), for example, 115 (105 + 10), after the sixth time After adjustment, node g will switch the optimal path to node a to path 16 (excluding link b→a), since the link cost change (105) of link b→a after the sixth adjustment is less than The switching threshold of the node b, therefore, the node b does not switch the optimal path to the node a to the path 1, thereby avoiding the node b and the node g due to the hardware capability of the node b and the processing environment being better than the node g. A microring phenomenon occurs between them. Moreover, since the node i has switched the optimal path to the node a to the path 22 after the fourth adjustment, the micro-ring phenomenon does not occur between the node g and the node i.

The link cost of the link b→a adjusted and released for the seventh time may be any value less than 130 (120 + 10) and greater than 120 (110 + 10), for example, 125 (115 + 10), after the seventh time After adjustment, node c will switch the optimal path to node a to path 4 (excluding link b→a), because the link cost change (115) of link b→a after the seventh adjustment is less than The switching threshold of the node b, therefore, the node b does not switch the optimal path to the node a to the path 1, thereby avoiding the node b and the node c due to the hardware capability of the node b and the processing environment being better than the node c. A microring phenomenon occurs between them. Moreover, since node d has switched the optimal path to node a to path 7 after the fifth adjustment, micro-ring phenomenon does not occur between node c and node d.

The link cost of link b→a, which is adjusted and released for the eighth time, can be any value greater than 130 (120 + 10), for example, 130 (125 + 10). After the eighth adjustment, node b will Node a The optimal path is switched to path 1 (excluding link b→a). Since node g has switched the optimal path to node a to path 16 after the sixth adjustment, node b and node g are not A microring phenomenon will occur. Moreover, since the node c has switched the optimal path to the node a to the path 4 after the seventh adjustment, the micro-ring phenomenon does not occur between the node b and the node c.

For example, optionally, the network device performs N times adjustment on the link cost of the first link according to the path cost change value of each target node, including:

When the first link fails, the network device performs a third sorting process on the path cost change value of each target node in an incremental manner;

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is greater than The fifth value is less than the sixth value, wherein the fifth value is the i-th path cost change value after the third sorting process, and the sixth value is the i-th path after the third sorting process The cost change value.

The link cost of the link b→a that is first adjusted and released may be any value less than 70 (60 + 10) and greater than 60 (50 + 10), for example, 65 ( 55 + 10 ), after the first time After adjustment, only node f will switch the optimal path to node a to path 13 (excluding link b→a), due to the change in link cost of link b→a after the first adjustment (55) Less than the switching threshold of node e, Therefore, the node e does not switch the optimal path to the node a to the path 10, so that the micro-ring phenomenon between the node e and the node f can be avoided due to the hardware capability of the node e and the processing environment being better than the node f.

The link cost of the second adjusted and released link b→a can be any value less than 80 (70+ 10) and greater than 70 (60 + 10), for example, 75 (65 + 10), after the second time After adjustment, node j will switch the optimal path to node a to path 25 (excluding link b→a), because the link cost change (65) of link b→a after the second adjustment is less than The switching threshold of the node i, therefore, the node i does not switch the optimal path to the node a to the path 110, thereby avoiding the node i and the node j due to the hardware capability of the node i and the processing environment being better than the node j A microring phenomenon occurs between them.

The link cost of link b→a, which is adjusted and released for the fourth time, can be any value less than 100 (90 + 10) and greater than 90 (80+ 10), for example, 95 (85 + 10), after the fourth time After adjustment, node i will switch the optimal path to node a to path 22 (excluding link b→a), because the link cost change (85) of link b→a after the fourth adjustment is less than The switching threshold of the node g, therefore, the node g does not switch the optimal path to the node a to the path 16, thereby avoiding the node g and the node i due to the hardware capability of the node g and the processing environment being better than the node i A microring phenomenon occurs between them. Moreover, since the node j has switched the optimal path to the node a to the path 25 after the second adjustment, the micro-ring phenomenon does not occur between the node i and the node j.

The link cost of link b→a, which is adjusted and released for the fifth time, can be any value less than 110 (100+ 10) and greater than 100 (90+ 10), for example, 105 (95 + 10), after the fifth time After adjustment, node d will switch the optimal path to node a to path 7 (excluding link b→a), because the link cost change (95) of link b→a after the fifth adjustment is less than The switching threshold of the node c, therefore, the node c does not switch the optimal path to the node a to the path 4, thereby avoiding the node c and the node d due to the hardware capability of the node c and the processing environment being better than the node d Micro-interval Ring phenomenon. Moreover, since the node e has switched the optimal path to the node a to the path 10 after the third adjustment, the micro-ring phenomenon does not occur between the node d and the node e.

The link cost of link b→a, which is adjusted and released for the sixth time, may be any value less than 120 (110 + 10) and greater than 110 (100 + 10), for example, 115 (105 + 10), after the sixth time After adjustment, node g will switch the optimal path to node a to path 16 (excluding link b→a), since the link cost change (105) of link b→a after the sixth adjustment is less than The switching threshold of the node b, therefore, the node b does not switch the optimal path to the node a to the path 1, thereby avoiding the node b and the node g due to the hardware capability of the node b and the processing environment being better than the node g. A microring phenomenon occurs between them. Moreover, since the node i has switched the optimal path to the node a to the path 22 after the fourth adjustment, the micro-ring phenomenon does not occur between the node g and the node i.

The link cost of link b→a, which is adjusted and released for the eighth time, can be any value greater than 130 (120 + 10), for example, 130 (125 + 10). After the eighth adjustment, node b will The optimal path of node a is switched to path 1 (excluding link b→a). Since node g has switched the optimal path to node a to path 16 after the sixth adjustment, node b and node g There is no microring phenomenon between them. Moreover, since the node c has switched the optimal path to the node a to the path 4 after the seventh adjustment, the microring phenomenon does not occur between the node b and the node c.

The above is an example of determining the number of adjustments according to the node to be processed, but the present invention is not limited thereto. For example, when the link b→a fails, two adjustments can be made to make two Or the previous hop node on the first path of the two or more pairs of pending nodes migrates the optimal path out of the first path before the next hop node. As an implementation method, the following actions can be performed.

The network device determines, according to the path cost change value of each node to be processed, an adjustment range of each to-be-processed node, where the adjustment range of a to-be-processed node is less than or equal to the path cost change value of the to-be-processed node, and is greater than or equal to the to-be-processed node. The path cost change value of the reference node of the processing node, where the reference node of the node to be processed is the node with the largest path cost change value of the previous hop node of each to-be-processed node in each first path;

△ cost ( i, max ) = D(i, max) - D ( i, min ).

Δ cost ( i, min ) =MAX{ Δ cost ( j , max ) } , where node j is the son node of node i under normal conditions of the first link (or node j is node i in the first path) The last hop node). Therefore, the range of cost values (ie, the adjustment range) of the first link that needs to be adjusted for the anti-micro ring switching of the node i is <厶. (^ ( i, min ) + K, A cost ( i, max ) + K>, where K is the cost value when the first link is normal.

Thereafter, adjustment can be made according to the adjustment range of each node. For example, in system 200, the adjustment range of node g is [80, 100], the adjustment range of node c is [90, 110], and node g constitutes The handover threshold of the node of the node pair to be processed is not between the handover threshold of the node c and the node d (that is, the handover threshold of the node b is greater than 110, and the handover threshold of the node i is less than 90), and the link overhead is adjusted. In order for node c to switch the optimal path, the node g can be switched at the same time.

As shown in FIG. 3, since the paths of the remaining nodes to the node b may be different, for example, the node node d, the node e, and the node f need to be switched in order from the hop count of the node a, and the node is sequentially switched. g, node i, node j also need to be switched in order from the hop count of node a from small to large. Therefore, for nodes respectively located on the above two paths, there may be a case of simultaneously switching to the optimal path by one overhead adjustment.

1. Initialize three lists: TentList, CandList, and OutPutList.

2. Link failure The nearest node (here, node b) is the first to do anti-micro ring switching, remember that the node is self.

3. Push the node self into the TentList.

4. Obtain a node x from the TentList, where the node x is the adjustment range in the TentList The node with the lowest lower limit. If the TentList is empty, the algorithm ends.

5. Push node X into OutPutList.

6. Put all the child nodes y (here, node c and node g) of node X that may have micro-ring switching into the CandList.

7. Determine the maximum switching threshold cost 1 of the switching thresholds of all nodes in the CandList, and the switching threshold cost 2 of the node z in the TentList. If cost 2> cost 1 , the node z is deleted from the TentList.

11. Jump execution 4.

Thus, it can be determined that each node in the OutPutList is the target node.

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is greater than The fifth value is less than the sixth value, wherein the sixth value is the i+1th path cost change value after the third sorting process.

The fifth value is the largest value among the path cost change values of the fifth target nodes, the fifth item The target node and the sixth target node form a pair of pending nodes, and the fifth target node is a previous hop node of the sixth target node in the first path, and the sixth target node is processed by the third sorting process. The node corresponding to the i + 1 path cost change value.

Specifically, after sorting the switching thresholds of the nodes in the OutPutList (here, in an ascending order as an example), T adjustments can be made, T being the number of nodes included in the OutPutList, as described above, In system 200, T is five.

That is, the link cost of the link b→a that is first adjusted and released may be any value less than 80 (70 + 10) and greater than 60 (50 + 10), for example, 61 ( 51 + 10 ), After an adjustment, node f will switch the optimal path to node a to path 13 (excluding link b→a), due to the change in link cost of link b→a after the first adjustment (51) ) is smaller than the switching threshold of the node e, therefore, the node e does not switch the optimal path to the node a to the path 10, thereby avoiding the node e and the node due to the hardware capability of the node e and the processing environment being better than the node f A microring phenomenon occurs between f.

It should be noted that, as shown in FIG. 3, in the system 200, the node e has only one child node, but the present invention is not limited thereto. When the node e has multiple child nodes, the adjusted The link cost of link b→a is greater than the largest value of the path cost change value of each child node of node e.

The link cost of the link b→a adjusted and released for the second time may be any value less than 100 (90 + 10) and greater than 80 (70 + 10), for example, 81 (71 + 10), after the second time After adjustment, node e will switch the optimal path to node a to path 10 (excluding link b→a), because the link cost change (71) of link b→a after the second adjustment is less than The switching threshold of the node d, therefore, the node d does not switch the optimal path to the node a to the path 7, thereby avoiding the node d and the node e due to the hardware capability of the node d and the processing environment being better than the node e A microring phenomenon occurs between them. Moreover, since the node f has switched the optimal path to the node a to the path 13 after the first adjustment, the micro-ring phenomenon does not occur between the node e and the node f. And, after the second adjustment, node j will switch the optimal path to node a to path 25 (excluding link b→a), due to the link of link b→a after the second adjustment. The cost change (71) is smaller than the switching threshold of the node i. Therefore, the node i does not switch the optimal path to the node a to the path 110, thereby avoiding that the hardware capability of the node i and the processing environment are better than the node j. A microring phenomenon occurs between node i and node j.

It should be noted that, as shown in FIG. 3, in the system 200, the node d has only one child node. However, the present invention is not limited thereto. When the node d has a plurality of child nodes, the link cost of the adjusted link b→a may be made larger than the maximum path cost change value of each child node of the node d. value.

The link cost of the link b→a adjusted and released for the third time may be any value less than 120 (110 + 10) and greater than 100 (90 + 10), for example, 101 (91 + 10), after the third time After adjustment, node d will switch the optimal path to node a to path 7 (excluding link b→a), because the link cost change (91) of link b→a after the third adjustment is less than The switching threshold of the node c, therefore, the node c does not switch the optimal path to the node a to the path 4, thereby avoiding the node c and the node d due to the hardware capability of the node c and the processing environment being better than the node d A microring phenomenon occurs between them. Moreover, since the node e has switched the optimal path to the node a to the path 10 after the second adjustment, the micro-ring phenomenon does not occur between the node d and the node e. And, after the second adjustment, node i will switch the optimal path to node a to path 22 (excluding link b→a), due to the link of link b→a after the second adjustment. The cost change (91) is smaller than the switching threshold of the node g. Therefore, the node g does not switch the optimal path to the node a to the path 16, thereby avoiding that the hardware capability of the node g and the processing environment are better than the node i. A microring phenomenon occurs between node g and node i. Moreover, since the node j has switched the optimal path to the node a to the path 25 after the second adjustment, the micro-ring phenomenon does not occur between the node i and the node j.

The link cost of the link b→a adjusted and released for the fourth time may be any value less than 130 (120 + 10) and greater than 120 (110 + 10), for example, 121 (111 + 10), after the fourth time After adjustment, node c will switch the optimal path to node a to path 4 (excluding link b→a), because the link cost change (115) of link b→a after the fourth adjustment is less than The switching threshold of the node b, therefore, the node b does not switch the optimal path to the node a to the path 1, thereby avoiding the node b and the node c due to the hardware capability of the node b and the processing environment being better than the node c. A microring phenomenon occurs between them. And, since the node d has switched the optimal path to the node a to the path 7 after the third adjustment, the micro-ring phenomenon does not occur between the node c and the node d. And, after the fourth adjustment, node g will switch the optimal path to node a to path 16 (excluding link b→a), due to the link of link b→a after the fourth adjustment. The overhead change (105) is smaller than the switch of node b Threshold, therefore, node b does not switch the optimal path to node a to path 1, thus avoiding the occurrence of micro-ring between node b and node g due to the hardware capability of node b and the processing environment being better than node g. phenomenon. And, since the node i has switched the optimal path to the node a to the path 22 after the third adjustment, the micro-ring phenomenon does not occur between the node g and the node i.

It should be noted that, as shown in FIG. 3, in the system 200, since the node c is the node with the largest path cost change value among the child nodes (node c and node g) of the node b, the link cost of the current adjustment is made. Greater than the path cost change value of node c.

The link cost of link b→a, which is adjusted and released for the fifth time, can be any value greater than 130 (120 + 10), for example, 130 (125 + 10). After the eighth adjustment, node b will The optimal path of node a is switched to path 1 (excluding link b→a). Since node g has switched the optimal path to node a to path 16 after the fourth adjustment, node b and node g There is no microring phenomenon between them. Moreover, since the node c has switched the optimal path to the node a to the path 4 after the fourth adjustment, the microring phenomenon does not occur between the node b and the node c.

For the same reason, the switching thresholds of the nodes in the OutPutList can be arranged in descending order and adjusted. The process is similar to the above process. Here, in order to avoid redundancy, the description thereof is omitted.

The network device adjusts the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is greater than The seventh value is less than the eighth value, wherein the eighth value is the N-th path cost change value after the fourth sorting process,

The seventh value is the largest value among the path cost change values of the seventh target nodes, and the seventh target node and the eighth target node form a pair of pending nodes, and the seventh target node is in the first path. a last hop node of the eighth target node, the eighth target node is processed after the fourth sorting The node corresponding to the Nth _i + 1 path cost change value.

Optionally, the network device performing at least two adjustments on the link cost of the first link includes: determining, by the network device, a processing time required for each target node to calculate an optimal path;

Determining, by the network device, a time interval between the at least two adjustments according to the processing time; the network device performs at least two adjustments on the link overhead of the first link according to the time interval, in the present invention In an embodiment, the required recomputation processing time of the node capable of completing the optimal path reselection in the current adjustment may also be determined. For example, hardware information of each node may be obtained (for example, from a vendor), according to the hardware. The information is estimated by the calculation capability, thereby determining the recalculation processing time, and determining the time of the next adjustment and release, so that the node that can complete the optimal path reselection in the current adjustment completes the reselection, and then performs the next adjustment and release. .

According to the method for adjusting the link overhead according to the embodiment of the present invention, the occurrence of the microring phenomenon can be further reliably avoided.

According to the method for adjusting the link cost, the link cost of the first link directly connected to the first node is adjusted multiple times, including the first link and the first link is normal. In each of the first paths that are the optimal links to the first node, when the first link recovers from the failure, the pair of nodes to be processed that may have a microring phenomenon may be on the first path. The one-hop node migrates the optimal path to the first node to the first path before the previous hop node, or in the pair of pending nodes that may cause the micro-ring phenomenon when the first link fails. The previous hop node on the path migrates the optimal path to the first node out of the first path before the next hop node, thereby preventing the micro ring phenomenon from occurring between the pairs of nodes to be processed.

The method for adjusting the link overhead according to the embodiment of the present invention is described in detail above with reference to FIG. 1 to FIG. 3. Hereinafter, an apparatus for adjusting link overhead according to an embodiment of the present invention will be described in detail with reference to FIG.

Figure 4 shows a schematic block diagram of an apparatus 300 for adjusting link overhead in accordance with an embodiment of the present invention. As shown in Figure 4, the apparatus 300 includes:

The to-be-processed node determining unit 310 is configured to determine at least one pair of to-be-processed nodes from at least three nodes when the first link fails or recovers from the fault, where each pair of to-be-processed nodes includes a chain Two pending nodes connected to the path, the to-be-processed node can pass without including The second path of the first link sends a message to the first node of the at least three nodes, and the to-be-processed node can pass the first path including the first link to the first link when the first link is normal. The first node sends a message, where a second path is a path from a to-be-processed node with a minimum total link overhead in a path to the first node that does not include the first link, where a first path is The optimal path from the node to be processed to the first node when the first link is normal, and the up-and-down relationship between the nodes to be processed in the pair of nodes to be processed on the second path and the first path Differentiating, the second node of the at least three nodes is directly connected to the first node by using the first link, where the first link is used to transmit data that needs to be sent to the first node;

The path cost change value determining unit 320 is configured to determine a path cost change value of each to-be-processed node, where a path cost change value of each to-be-processed node is a difference between a first path cost and a second path cost of each to-be-processed node, The first path cost is a total link cost on the first path when the first link is normal, and the second path cost is a total link cost on the second path;

The link cost adjustment unit 330 is configured to adjust the link cost of the first link at least twice according to the path cost change value of each to-be-processed node, so that when the first link recovers from the fault, Having the next hop node on the first path of each pair of pending nodes migrate the optimal path to the first node to the first path prior to the previous hop node, or

When the first link fails, the previous hop node on the first path of each pair of pending nodes migrates the optimal path to the first node out of the first path before the next hop node.

Optionally, the link cost adjustment unit 330 is further configured to determine, from the to-be-processed node, N target nodes, where the number of the target nodes is less than or equal to the number of the to-be-processed nodes;

And adjusting, according to the path cost change value of each target node, the link cost of the first link to be adjusted N times.

Optionally, the link cost adjustment unit 330 is specifically configured to use all of the to-be-processed nodes as the N target nodes.

Optionally, the link cost adjustment unit 330 is configured to: when the first link recovers from the fault, the network device performs a first sorting process on the path cost change value of each target node in a decreasing manner. ;

The N-time adjustment is performed on the link cost of the first link, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is smaller than the first The value is greater than the second value, where the first value is the i-th path cost change value after the first sorting process, and the second value is the i-th path cost after the first sorting process Change value. Optionally, the link cost adjustment unit 330 is configured to: when the first link recovers from the fault, the network device performs a second sorting process on the path cost change value of each target node in an incremental manner. ;

The N-time adjustment is performed on the link cost of the first link, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is smaller than the third. The value is greater than the fourth value, wherein the third value is the N-th + 1 path cost change value after the second sorting process, and the fourth value is the N-th after the second sorting process i path cost change values.

Optionally, the link cost adjustment unit 330 is configured to: when the first link fails, the network device performs a third sorting process on the path cost change value of each target node in an incremental manner;

The N-time adjustment is performed on the link cost of the first link, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is greater than the fifth. The value is less than the sixth value, wherein the sixth value is the i-th path cost change value after the third sorting process, and the fifth value is the i-th path cost after the third sorting process Change value.

Optionally, the link cost adjustment unit 330 is configured to: when the first link fails, the network device performs a fourth sorting process on the path cost change value of each target node in a decreasing manner;

The N-time adjustment is performed on the link cost of the first link, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is greater than the seventh. The value is less than the eighth value, wherein the eighth value is the Nth-th path cost change value after the fourth sorting process, and the seventh value is the N-th+ after the fourth sorting process 1 path cost change value.

Optionally, the link cost adjustment unit 330 is further configured to determine a processing time required for each target node to calculate an optimal path;

For determining a time interval between the at least two adjustments according to the processing time;

And performing, according to the time interval, performing at least two adjustments on a link cost of the first link. Optionally, the device is the second node.

Optionally, the link cost adjustment unit 330 is specifically configured to determine, according to a path cost change value of each to-be-processed node, an adjustment range of each to-be-processed node, where an adjustment range of a to-be-processed node is less than or equal to the to-be-processed node. The path cost change value is greater than or equal to the path cost change value of the reference node of the to-be-processed node, and the reference node of the to-be-processed node is the path cost change value of the to-be-processed node in the previous hop node on each first path. Largest node And configured to perform at least two adjustments on the link cost of the first link according to the adjustment range of the to-be-processed node.

The apparatus 300 for adjusting the link overhead according to the embodiment of the present invention may correspond to a network device (for example, a second node) in the method of the embodiment of the present invention, and each unit in the apparatus 300 is a module and the foregoing other operations and In order to implement the corresponding process of the method 100 in FIG. 1 , the functions are not described here.

The device for adjusting the link cost according to the embodiment of the present invention, by adjusting the link cost of the first link directly connected to the first node, including the first link and being normal on the first link In each of the first paths that are the optimal links to the first node, when the first link recovers from the failure, the pair of nodes to be processed that may have a microring phenomenon may be on the first path. The one-hop node migrates the optimal path to the first node to the first path before the previous hop node, or in the pair of pending nodes that may cause the micro-ring phenomenon when the first link fails. The previous hop node on the path migrates the optimal path to the first node out of the first path before the next hop node, thereby preventing the micro ring phenomenon from occurring between the pairs of nodes to be processed.

The method for adjusting the link cost according to the embodiment of the present invention is described in detail above with reference to FIG. 1 to FIG. 3. Hereinafter, the device for adjusting the link overhead according to the embodiment of the present invention is described in detail with reference to FIG.

FIG. 5 shows a schematic block diagram of a device 400 for adjusting link overhead in accordance with an embodiment of the present invention. As shown in FIG. 5, the device 400 includes:

Bus 410;

a processor 420 connected to the bus;

a memory 430 connected to the bus;

The processor 420 calls, by using the bus 410, a program stored in the memory 430, where the network device determines at least one of the at least three nodes when the first link fails or recovers from a failure. a pair of nodes to be processed, wherein each pair of nodes to be processed includes two nodes to be processed connected via a link, and the node to be processed can send a report to the first node through a second path that does not include the first link. And the node to be processed can send a message to the first node by using a first path including the first link when the first link is normal, where a second path is from a to-be-processed node. a path to the first node that does not include the first link in which the total link overhead is the smallest, and a first path is the most from a pending node to the first node when the first link is normal. An optimal path, and each of the to-be-processed nodes in each pair of pending nodes has a different up-and-down relationship on the second path and the first path; a path cost change value of each node to be processed, where a path cost change value of each to-be-processed node is a difference between a first path cost of each to-be-processed node and a second path cost, where the first path cost is The total link cost on the first path when a link is normal, and the second path cost is the total link overhead on the second path;

The link cost of the first link is adjusted at least twice according to the path cost change value of each node to be processed, so that each of the to-be-processed nodes is centered when the first link recovers from the fault. The next hop node on the first path migrates the optimal path to the first node to the first path prior to the previous hop node, or

Optionally, the processor 420 is further configured to determine, from the to-be-processed node, N target nodes, where the number of the target nodes is less than or equal to the number of the to-be-processed nodes;

Optionally, the processor 420 is specifically configured to use all of the to-be-processed nodes as the N target nodes.

Optionally, the processor 420 is configured to: when the first link recovers from the fault, the network device performs a first sorting process on the path cost change value of each target node in a decreasing manner;

The N-time adjustment is performed on the link cost of the first link, so that the difference between the link cost of the first link after the i-th adjustment and the link cost when the first link is normal is smaller than the first The value is greater than the second value, where the first value is the i-th path cost change value after the first sorting process, and the second value is the i-th path cost after the first sorting process Change value.

Optionally, the processor 420 is configured to: when the first link recovers from the fault, the network device performs a second sorting process on the path cost change value of each target node in an incremental manner;

Optionally, the processor 420 is specifically configured to: when the first link fails, the network device Performing a third sorting process on the path cost change value of each target node in an incremental manner; and performing N times adjustment on the link cost of the first link, so that the first link after the ith adjustment is performed The difference between the link cost and the link cost when the first link is normal is greater than a fifth value and less than a sixth value, wherein the sixth value is the i+1th path cost after the third sorting process The change value, the fifth value is the i-th path cost change value after the third sorting process.

Optionally, the processor 420 is specifically configured to: when the first link fails, the network device performs a fourth sorting process on the path cost change value of each target node in a decreasing manner;

Optionally, the processor 420 is further configured to determine a processing time required for each target node to calculate an optimal path;

And performing, according to the time interval, performing at least two adjustments on a link cost of the first link. Optionally, the device 400 is the second node.

Optionally, the processor 420 is specifically configured to determine, according to a path cost change value of each node to be processed, an adjustment range of each to-be-processed node, where an adjustment range of a to-be-processed node is a path smaller than or equal to the to-be-processed node. The cost change value is greater than or equal to the path cost change value of the reference node of the to-be-processed node, and the reference node of the to-be-processed node is the path cost change value of the to-be-processed node in the previous hop node on each first path. Largest node

And adjusting, according to the adjustment range of the to-be-processed node, the link overhead of the first link to be adjusted at least twice.

The device 400 for adjusting the link overhead according to the embodiment of the present invention may correspond to a network device (for example, a second node) in the method of the embodiment of the present invention, and each unit in the device 400 is a module and the foregoing other operations and In order to implement the corresponding process of the method 100 in FIG. 1 , the functions are not described here.

It should be understood that, in the embodiment of the present invention, the processor 420 may be a central processing unit (a central processing unit), and the processor 420 may also be another general-purpose processor, a digital signal processor (DSP). ), application specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and the like. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like. The memory 430 can include read only memory and random access memory and provides instructions and data to the processor 610. A portion of the memory 430 may also include a non-volatile random access memory. For example, the memory 430 can also store information of the device type.

The bus 410 can include, in addition to the data bus, a power bus, a control bus, and a status signal bus. However, for clarity of description, various buses are labeled as bus system 410 in the figure. The logic circuit or the instruction in the form of software is completed. The steps of the method disclosed in the embodiments of the present invention may be directly implemented as a hardware processor, or may be performed by a combination of hardware and software modules in the processor. The software modules can be located in random memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, etc., which are well established in the art. The storage medium is located in the memory 430. The processor 420 reads the information in the memory 430 and combines the hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

The device for adjusting the link cost according to the embodiment of the present invention performs the multiple adjustments on the link cost of the first link directly connected to the first node, including the first link and the first link is normal. In each of the first paths that are the optimal links to the first node, when the first link recovers from the failure, the pair of nodes to be processed that may have a microring phenomenon may be on the first path. The one-hop node migrates the optimal path to the first node to the first path before the previous hop node, or in the pair of pending nodes that may cause the micro-ring phenomenon when the first link fails. The previous hop node on the path migrates the optimal path to the first node out of the first path before the next hop node, thereby preventing the micro ring phenomenon from occurring between the pairs of nodes to be processed.

It should be understood that the term "and/or" in this context is merely an association describing the associated object, indicating that there may be three relationships, for example, A and / or B, which may represent: A exists separately, and A and B exist simultaneously There are three cases of B alone. In addition, the character "/" in this article generally indicates that the context object is an "or" relationship.

It should be understood that, in various embodiments of the present invention, the size of the sequence numbers of the above processes does not mean the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not be taken to the embodiments of the present invention. The implementation process constitutes any limitation.

Those skilled in the art will appreciate that the elements and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, or in computer software and electronic hardware. Come together to achieve. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods to implement the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present invention.

It will be apparent to those skilled in the art that, for the convenience of the description and the cleaning process, the specific operation of the system, the device and the unit described above may be referred to the corresponding processes in the foregoing method embodiments, and details are not described herein again.

In the several embodiments provided herein, it should be understood that the disclosed systems, devices, and methods may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be electrical, mechanical or otherwise.

The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention, which is essential to the prior art or part of the technical solution, may be embodied in the form of a software product stored in a storage medium, including The instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program codes. .

The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto, and any person skilled in the art can easily within the technical scope disclosed by the present invention. Any changes or substitutions are contemplated as being within the scope of the invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.

Claims

Rights request

1. A method for adjusting link overhead, characterized in that it is executed in a communication system including at least three nodes, wherein the first node and the second node are directly connected through a first link, and the first node The link is used to transmit data that needs to be sent to the first node. The method includes: when the first link fails or recovers from the failure, the network device determines at least one of the at least three nodes. A pair of nodes to be processed, wherein each pair of nodes to be processed includes two nodes to be processed connected via a link, and the node to be processed can communicate with the first node through a second path that does not include the first link. The node sends a message, and the node to be processed can send a message to the first node through a first path including the first link when the first link is normal, where a second path is The path from a node to be processed to the first node that has the smallest total link cost among the paths excluding the first link. A first path is from a node to be processed when the first link is normal. The optimal path from the processing node to the first node, and the up-and-down hop relationships of the nodes to be processed in each pair of nodes to be processed on the second path and the first path are different; the network device determines each The change value of the path cost of the node to be processed. The change value of the path cost of each node to be processed is the difference between the first path cost and the second path cost of each node to be processed. The first path cost is when the first link The total link cost on the first path when the road is normal, and the second path cost is the total link cost on the second path;

The network device adjusts the link cost of the first link at least twice according to the path cost change value of the node to be processed, so that when the first link recovers from a fault, each The next hop node on the first path in the pair of nodes to be processed migrates the optimal path to the first node to the first path before the previous hop node, or

When the first link fails, the previous hop node on the first path in each node pair to be processed is moved out of the optimal path to the first node before the next hop node. First path.

2. The method according to claim 1, wherein the network device adjusts the link cost of the first link at least twice according to the path cost change value of the node to be processed, including:

The network device determines the adjustment range of each node to be processed based on the path cost change value of each node to be processed, wherein the adjustment range of a node to be processed is less than or equal to the path cost change value of the node to be processed, and is greater than or equal to The path cost change value of the reference node of the node to be processed. The reference node of a node to be processed is the previous hop of the node to be processed on each first path. The node with the largest path cost change value among the nodes;

The network device adjusts the link overhead of the first link at least twice according to the adjustment range of the node to be processed.

3. The method according to claim 1 or 2, characterized in that, the network device adjusts the link cost of the first link at least twice according to the path cost change value of the node to be processed, include:

The network device determines N target nodes from the nodes to be processed, where the number of target nodes is less than or equal to the number of nodes to be processed;

4. The method according to claim 3, characterized in that the network device determines N target nodes from the nodes to be processed, including:

The network device uses all the nodes to be processed as the N target nodes.

5. The method according to claim 3 or 4, characterized in that, the network device adjusts the link cost of the first link N times according to the path cost change value of each target node, including :

When the first link recovers from a failure, the network device performs a first sorting process on the path cost change values of each of the target nodes in a decreasing manner;

The network device adjusts the link cost of the first link N times, so that the link cost of the first link after the i-th adjustment is equal to the link cost when the first link is normal. The difference is smaller than the first value and larger than the second value, where the first value is the i-th path cost change value after the first sorting process, and the second value is the i-th path cost change value after the first sorting process. The i + 1th path cost change value.

6. The method according to claim 3 or 4, characterized in that, the network device adjusts the link cost of the first link N times according to the path cost change value of each target node, including :

When the first link recovers from a failure, the network device performs a second sorting process on the path cost change values of each of the target nodes in an incremental manner;

The network device adjusts the link cost of the first link N times, so that the link cost of the first link after the i-th adjustment is equal to the link cost when the first link is normal. The difference is less than the third value and greater than the fourth value, wherein the third value is the Nth - after the second sorting process i + 1 path cost change values, and the fourth value is the N-ith path cost change value after the second sorting process.

7. The method according to claim 3 or 4, characterized in that, the network device adjusts the link cost of the first link N times according to the path cost change value of each target node, including :

When the first link fails, the network device performs a third sorting process on the path cost change values of each of the target nodes in an incremental manner;

The network device adjusts the link cost of the first link N times, so that the link cost of the first link after the i-th adjustment is equal to the link cost when the first link is normal. The difference is greater than the fifth value and less than the sixth value, wherein the sixth value is the i + 1th path cost change value after the third sorting process, and the fifth value is the change value after the third sorting process. The processed i-th path cost change value.

8. The method according to claim 3 or 4, characterized in that, the network device adjusts the link cost of the first link N times according to the path cost change value of each target node, including :

When the first link fails, the network device performs a fourth sorting process on the path cost change values of each of the target nodes in a decreasing manner;

The network device adjusts the link cost of the first link N times, so that the link cost of the first link after the i-th adjustment is equal to the link cost when the first link is normal. The difference is greater than the seventh value and less than the eighth value, wherein the eighth value is the N-ith path cost change value after the fourth sorting process, and the seventh value is the N-th path cost change value after the fourth sorting process. The processed N_i+1th path cost change value.

9. The method according to any one of claims 1 to 8, wherein the network device adjusts the link overhead of the first link at least twice including:

The network device determines the processing time required for each target node to calculate the optimal path; the network device determines the time interval between at least two adjustments based on the processing time;

The network device adjusts the link overhead of the first link at least twice according to the time interval.

10. The method according to any one of claims 1 to 9, characterized in that the network device is the second node.

11. A device for adjusting link overhead, characterized in that the device includes: a node determination unit to be processed, configured to select a node from at least three nodes when the first link fails or recovers from a failure. Determine at least one node pair to be processed, wherein each node pair to be processed includes two nodes to be processed connected via a link, and the node to be processed can be connected to the at least one node through a second path that does not include the first link. The first node of the three nodes sends a message, and the node to be processed can send a message to the first node through the first path including the first link when the first link is normal, where , a second path is the path from a node to be processed to the first node that has the smallest total link cost among the paths excluding the first link, and a first path is the path between the first link and the node to be processed. The optimal path from a node to be processed to the first node when the road is normal, and the up and down hop relationships of the nodes to be processed in each pair of nodes to be processed on the second path and the first path are different, The second node among the at least three nodes is directly connected to the first node through the first link, and the first link is used to transmit data that needs to be sent to the first node;

The path cost change value determination unit is used to determine the path cost change value of each node to be processed, where the path cost change value of each node to be processed is the difference between the first path cost and the second path cost of each node to be processed, The first path cost is the total link cost on the first path when the first link is normal, and the second path cost is the total link cost on the second path;

A link cost adjustment unit, configured to adjust the link cost of the first link at least twice according to the path cost change value of each node to be processed, so that when the first link recovers from a failure, , so that the next hop node on the first path in each node pair to be processed migrates the optimal path to the first node to the first path before the previous hop node, or

When the first link fails, the previous hop node on the first path in each node pair to be processed is migrated out of the optimal path to the first node before the next hop node. First path.

12. The method according to claim 11, characterized in that, the link cost adjustment unit is specifically configured to determine the adjustment range of each node to be processed according to the path cost change value of each node to be processed, wherein, one to be processed The adjustment range of a node is less than or equal to the path cost change value of the node to be processed, and greater than or equal to the path cost change value of the reference node of the node to be processed. The reference node of a node to be processed is the node to be processed in each The node with the largest path cost change value among the previous hop nodes on the first path;

For adjusting the link overhead of the first link according to the adjustment range of the node to be processed. Make at least two adjustments.

13. The device according to claim 11 or 12, wherein the link cost adjustment unit is further configured to determine N target nodes from the nodes to be processed, wherein the number of target nodes is less than Equal to the number of nodes to be processed;

For adjusting the link cost of the first link N times according to the change value of the path cost of each of the target nodes.

14. The device according to claim 13, wherein the link cost adjustment unit is specifically configured to use all of the nodes to be processed as the N target nodes.

15. The device according to claim 13 or 14, characterized in that, the link overhead adjustment unit is specifically configured to: when the first link recovers from a fault, the network device decreases the Perform a first sorting process on the path cost change values of each of the target nodes;

Used to adjust the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost of the first link when it is normal is less than The first value is greater than the second value, wherein the first value is the i-th path cost change value after the first sorting process, and the second value is the i-th path cost change value after the first sorting process. i + 1 path cost change value.

16. The device according to claim 13 or 14, characterized in that, the link overhead adjustment unit is specifically configured to, when the first link recovers from a fault, the network device in an incremental manner, Perform a second sorting process on the path cost change values of each target node;

Used to adjust the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost of the first link when it is normal is less than The third value is greater than the fourth value, wherein the third value is the N_i+1th path cost change value after the second sorting process, and the fourth value is the N_i+1th path cost change value after the second sorting process. The processed N-ith path cost change value.

17. The device according to claim 13 or 14, wherein the link overhead adjustment unit is specifically configured to adjust the network equipment to each location in an incremental manner when the first link fails. The path cost change value of the target node is subjected to the third sorting process;

Used to adjust the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost of the first link when it is normal is greater than The fifth value is less than the sixth value, wherein the sixth value is the i + 1th path cost change value after the third sorting process, and the fifth value is the i+1th path cost change value after the third sorting process. The i-th path cost change value.

18. The device according to claim 13 or 14, wherein the link overhead adjustment unit is specifically configured to adjust the network equipment to each location in a decreasing manner when the first link fails. The path cost change value of the target node is subjected to the fourth sorting process;

Used to adjust the link cost of the first link N times, so that the difference between the link cost of the first link after the i-th adjustment and the link cost of the first link when it is normal is greater than The seventh value is less than the eighth value, where the eighth value is the N-ith path cost change value after the fourth sorting process, and the seventh value is the N-ith path cost change value after the fourth sorting process. The N_i+1th path cost change value.

19. The device according to any one of claims 11 to 18, characterized in that the link cost adjustment unit is also used to determine the processing time required for each target node to calculate the optimal path; Determine the time interval between the at least two adjustments according to the processing time; and adjust the link overhead of the first link at least twice according to the time interval.

20. The device according to any one of claims 11 to 19, characterized in that the device is the second node.