JP2007049336A - Method for restoring fault, communication node and network - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the restoration ratio of paths during the restoration of multiple faults in a communication network by increasing restorable fault cases while securing a high-speed property equivalent to a dedicated protection shared restoration system. <P>SOLUTION: When gradually applying a plurality of fault restoration systems, many paths as possible are restored by a high-speed dedicated protection shared restoration system, and when there are paths which can not be restored, an aiding system is transited to a high-adaptive dynamic design type restoration system. Further, a policy designed so that the number of paths aided in succeeding restoration is increased is applied to preceding restoration to perform restoration processing. Further a high-speed path end switching system in which a fault position evaluation is simple is applied to the preceding fault restoration system, and in the succeeding restoration, paths can be switched at an optional switching point to perform flexible route setting and a switching control point between the preceding restoration system and the succeeding restoration system is changed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention is used for recovery from a path failure in a network.

  In a backbone network having a large path capacity in which traffic is aggregated in the network, service interruption due to a path failure has a great influence on the user. Therefore, recovery in a very short time is required for failure recovery. In addition, when a plurality of paths fail, it is necessary to restore as many paths as possible to increase the recovery rate.

  Conventional failure recovery technologies include a pre-designed recovery method that sets a backup path in advance of a failure, and a dynamic design type recovery method that sets an alternative path immediately after a failure occurs. The latter is excellent in adaptability.

  The dynamic design type restoration method evaluates the failure location after the failure occurs, assigns the route, and calculates the wavelength. Since time is required for the failure location assessment, route assignment, and wavelength calculation, the backbone network is used. Cannot meet the demand for high-speed recovery, and the applicable service class is greatly limited. For this reason, technologies capable of high-speed restoration such as Dedicated protection and shared restoration (for example, see Non-Patent Document 1), which is a pre-design type restoration that designs an alternative path before occurrence of a failure, have been devised.

  In these techniques, a backup path for the working path is designed in advance, and when a failure occurs, an alternative path for the failed path is set in the backup path determined by the design. In this method, the detour route and the allocated resource are determined in advance, so that extremely fast recovery is possible. However, it is unrealistic from the viewpoint of cost to allocate dedicated spare resources to many routes and manage the information assuming multiple failures, so currently only a single failure is assumed. And is implemented for the purpose of 100% recovery from a single failure.

  In this way, design is performed only in the case of failure assumed in advance, so for the failure that is not assumed in the design, there is topology connectivity that can be used for recovery in the event of an unexpected multiple failure, and Even when there are available resources, there are cases where recovery cannot be performed because only a predetermined route can be selected.

K. shimano, A .; Sahara, M .; Koga, Y. et al. Takagawa, and K.K. Sato: “Demonstration of Fast Restoration for Distributed Control Plane on Photonic Network,” ECOC2002, Copenhagen, Denmark, Sept. 2002. IETF RFC 3209. IETF RFC 3480. IETF RFC 2328. IETF RFC 1771. IETF RFC 1142.

  In the conventional pre-design type recovery technology, there are cases where recovery by the failure recovery technology is impossible for multiple failures due to system limitations. Moreover, it is difficult to satisfy the requirements for the recovery speed in the dynamic design type recovery method.

  The problem to be solved by the present invention is to increase the number of failure cases that can be recovered by improving or supplementing the conventional method while ensuring the same high speed as that of the prior-designed recovery method, which is the prior art. It is to improve the recovery rate of the path at the time of recovery.

  The present invention provides a function for exchanging control messages between communication nodes, a network path setting function according to a policy, a function for setting an alternative path according to a policy in the event of a path failure, and a policy for setting a path and an alternative path. This is a failure recovery method implemented by a communication node having a storage means, a management node having an information exchange means for a path accommodation design function and a communication node, and a network including these devices.

  The failure recovery process automatically detects a failure when a failure occurs in the path, and each node exchanges signaling protocol messages and communicates to perform autonomous and distributed failure recovery processing. . This recovery process is triggered by a failure detected by the communication node or a message exchanged between the communication nodes.

  This method applies multiple failure recovery methods step by step according to a preset scenario, and at that time, pre-designed recovery that enables high-speed processing by performing switch settings and resource reservation in advance. By restoring as many paths as possible with the method (see Non-Patent Document 1), and when there are paths that cannot be restored, the recovery method is transitioned to a more adaptable dynamic design type restoration method. While ensuring high speed, adaptability can be improved and the recovery rate can be improved.

  Further, in the present invention, by changing the switching control point between the restoration methods of the former stage and the latter stage, the flexibility of route search is improved, the number of discovered routes is increased, and the restoration rate is improved. Further, during the recovery process according to the failure recovery scenario, at a stage where high speed is important, high speed recovery is performed by selecting a switching control point capable of high speed recovery.

  1 and 2 show an example of the outline of the operation flow of the present invention. FIG. 1 is a flowchart showing a procedure at the start of path operation. FIG. 2 is a flowchart showing the procedure of failure recovery processing. In the present invention, at the start of operation, paths are accommodated and designed according to a preset policy for the working path and the prior path as shown in FIG. Subsequently, the working path and the preset path backup path are set by signaling according to the result of the accommodation design.

  When a failure occurs, use the first failure recovery method to restore the failed path (primary recovery) as shown in Fig. 2. If there is a path that cannot be recovered by the primary recovery, the recovery method is changed according to the scenario, and another recovery is performed. Perform recovery (secondary recovery) according to the policy using this method.

  By applying a high-speed pre-designed recovery method for primary recovery, and using a dynamic recovery method that can recover failures more adaptively than primary recovery for secondary recovery, the path recovery rate is improved while maintaining high speed. Can be made. In the present invention, a management node that manages the entire network is provided, and a backup path accommodation design is possible by applying the working path and the advance reservation type recovery method at the management node. Further, when performing failure recovery at the communication node, it is possible to perform recovery by applying the following policy.

Hereinafter, a policy for improving the recovery rate in the secondary recovery will be described. Policies a) and b) are policies for improving the recovery rate, and policy c) is a policy for speeding up.
a) When a plurality of failed paths are generated and a plurality of alternative paths are set from a plurality of grounds, if the alternative paths pass through the same link and resource contention occurs between the failed paths, a failure occurs. Compare the number of node independent paths q that can be set from the start node to the end node of the paths, or the sum Q of resources that can be set with q independent paths, and restore the failure path with priority q or Q with priority. Do.

  A description of Q and q is shown in FIG. FIG. 3 is a diagram illustrating a node-independent path between Node-1 and Node-6 and its resources when Node-1 is the start point and Node-6 is the end point. The network in FIG. 3 is a network composed of nodes from Node-1 to Node-6 and links from Link-1 to Link-7. Moreover, the resource which the link has was attached to all the links. Here, when Node-1 is the start point and Node-6 is the end point, if maximum node-independent routes that do not share nodes are secured from Node-1 to Node-6, three routes S1, S2, and S3 are obtained. It can be secured. In this case, q = 3 from Node-1 to Node-6.

  S1, S2, and S3 are configured by the following links.

S1: Link1 → Link2 → Link-3
S2: Link-4
S3: Link-5 → Link-6 → Link-7
Among the links constituting S1, S2, and S3, the minimum resources are as follows.

S1: 1
S2: 3
S3: 2
In this case, Q is the sum of these three values and is expressed as Q = 6.

For each path, the number of independent paths from the start node to the end node is calculated and held in advance by the communication node. In the example of FIG. 3, the number of independent paths is three, S1, S2, and S3. When a scramble occurs, the scramble is avoided by resetting an alternative path of a path having a small number of independent routes to a different path. A path having a large number of independent paths has a high probability that an alternative path can be set for another path. Therefore, the recovery rate can be increased by performing the above.
b) When performing secondary recovery, a link with a low resource share is preferentially selected to set an alternative path.

  The resource share rate Rshare is a different value for each link. In a certain link i, λi is a reserved backup resource allocated for a backup path by sharing resources, and Bk is a sum of resources of n backup paths passing through the link i when the resource is not shared. Is expressed by the following formula.

故障パスを復旧するために代替パスを設定する際、同時発生した他の故障パスの復旧のために既にリソースが占有されている可能性がある。このような状況の発生確率を抑えるために、リソースシェア率の低い予備リソースを選択して代替パスを設定する。

When an alternative path is set up to recover a failed path, there is a possibility that resources are already occupied for recovery of other failed paths that occurred at the same time. In order to suppress the occurrence probability of such a situation, an alternative path is set by selecting a spare resource having a low resource share rate.

A reserve resource with a higher resource share rate is more frequently used due to a failure. Therefore, avoiding links with a high resource share ratio for dynamic recovery, and selecting links with a low resource share ratio in preference and setting an alternative path increases the probability of occurrence of a situation in which spare resources are competing with other failed paths. Avoiding and reducing path setup failure during recovery.
c) When performing secondary recovery, make the best use of the pre-designed path with reserved resources. In secondary recovery, when setting an alternative path, partially use an available section other than the failure section of the pre-designed path that was attempted to be used in the primary recovery, and perform a new route search and resource allocation in the secondary recovery. Minimize the scope to do. In this way, a part of the route setting process for which resource reservation has been completed can be omitted, and the path setting can be speeded up.

  Further, the present invention provides a means for calculating a resource share rate for each link, a means for calculating the number q of independent paths between the path start point and the end point, and a resource share rate in order to perform failure recovery using the above policy. And means for recording the value of q.

  The resource share rate is determined using resource information collected by a routing protocol and a signaling message. Resources that are physically implemented are registered in the resource information database in the communication node and in the network management database in the management node.

  When setting the working path and protection path, the resources assigned to the working path and the resources assigned to the protection path are read from the signaling message, and the resource information database and the network management database are updated accordingly. The share rate is calculated using the information in this database.

  Further, the topology information collected by the routing protocol is recorded as a graph representing the topology in the route information database and the network management database. Using this graph, the values of q and Q are calculated by the route calculation function unit or the accommodation design engine.

  By applying the above policy and performing path accommodation design and restoration processing, it is possible to maximize the degree of improvement in the restoration rate by changing the method.

  That is, the first aspect of the present invention is configured by a plurality of communication nodes and a link connecting these communication nodes, and is applied to a network that performs communication by setting a path between any of the communication nodes. This is a failure recovery method in which when a failure occurs in the working path, the communication node sets an alternative path for the working path.

  Here, a feature of the present invention is that a failure recovery scenario in which a plurality of failure recovery procedures having different path setting methods or resource reservation methods and application orders of the plurality of failure recovery procedures are set in the communication node. Is stored in advance, and when a failure occurs, the communication node first performs one failure recovery procedure according to the failure recovery scenario, and if the failure recovery procedure cannot set an alternative path, A failure recovery procedure different from the failure recovery procedure initially performed by the setting method or the resource reservation method is to be performed next.

  This makes it possible to perform path recovery that meets requirements such as recovery rate and high speed for each path. In addition, failure information can be collected at the time of the previous stage failure, and by using it for subsequent stage recovery, more efficient route selection and resource allocation than when using a single recovery method. This makes it possible to perform recovery processing with a high recovery rate and high-speed recovery processing even for failure cases such as double failures that cannot be dealt with when a single recovery method is applied. Furthermore, it is possible to perform priority control of resource allocation to an efficient recovery path that could not be achieved when using a single recovery method.

  The message exchanged between the communication nodes in the failure recovery procedure includes information on whether or not a failure recovery procedure different in path setting method or resource reservation method from the first failure recovery procedure is to be executed next. be able to. As a result, each communication node can advance in advance the preparations necessary for performing the next different failure recovery procedure.

  Further, in the message exchanged between the communication nodes in the failure recovery procedure, the information indicating that the failure recovery procedure to be executed next is different from the first failure recovery procedure, and the failure recovery procedure differs in the resource reservation method. Information and information related to the recovery operation can be included.

  As a result, each communication node can perform more adaptive route calculation by using the received information, and include the information of the path that needs to be restored in the signaling message of the alternative path to be newly set. The necessary information can be transmitted to the communication node to set the alternative path. This information can be used for priority control of resource allocation and route calculation in a communication node in which an alternative path is newly set.

  Here, the failure information and the information related to the recovery operation are the failed link ID, interface ID, control network link ID, communication node ID connected to the failed link, the number of failed paths, and recoverable paths. , Their identifiers, bandwidth, IP address and network address of the start and end communication nodes, route and route setting method, transit area, path setting policy, path priority, command for switching, etc. It is.

  For example, the plurality of failure recovery procedures include a pre-design type failure recovery procedure for performing alternative path route calculation before failure occurrence and a dynamic design type failure recovery procedure for performing alternative path route calculation after failure occurrence. .

  Since this pre-design type recovery procedure performs path calculation and resource allocation calculation in advance, the recovery process can be significantly faster than the dynamic design type recovery procedure. Therefore, by providing a scenario for recovery using a pre-designed recovery procedure, it is possible to significantly reduce the recovery time of a fault path.

  In the present invention, for a single failure having a relatively high failure occurrence probability, high-speed failure recovery is performed by a pre-designed recovery procedure. On the other hand, for multiple faults with relatively low fault occurrence probabilities, fault recovery is carried out by adapting to the fault occurrence state each time using a dynamic design type recovery procedure.

  In this way, for failure scenarios with a high probability of failure, high-speed failure recovery is attempted using a pre-designed recovery procedure, while for scenarios with a low failure probability, failure recovery is performed using a dynamic design recovery procedure. By implementing the above, it is possible to increase the probability of successful failure relief while reducing the pre-design load required in the pre-design type recovery procedure.

  Since the pre-design type recovery method performs path calculation and resource allocation calculation in advance, it can perform recovery processing significantly faster than the dynamic design type recovery method. By providing recovery scenarios that can be recovered using a pre-designed recovery method for paths that can be recovered using a pre-designed recovery method, the average recovery time for failed paths can be significantly reduced. .

  In addition, the plurality of failure recovery procedures include a failure recovery procedure for switching the entire working path as a failure section to an alternate path at both ends of the working path, and a part of the working path as a failure section and switching only the failure section to another route. And a failure recovery procedure as an alternative path. As a result, it is possible to adaptively selectively use a roughly high-speed failure recovery procedure and a failure recovery procedure that can reduce network speed but reduce network redundancy.

  In addition, when a failure occurs in the link, the communication node or the management node holds information on the number of active paths that become a failure and the number of paths that can be set as an alternative path. If the number of working paths that become faulty is larger than the number of paths that can be set as an alternative path, it is possible to select a working path that sets an alternative path based on the policy set in advance in the communication node or management node. .

  As a result, each communication node or management node recovers autonomously based on a predetermined policy even when the number of working paths that fail is larger than the number of paths that can be set as alternative paths. A path can be selected. As a result, it is possible to avoid unnecessary signaling message transmission and path setting processing, and to reduce the burden on the communication node.

  Furthermore, by eliminating useless signaling messages, unnecessary resource contention can be avoided, and as a result, the time required to complete recovery processing for all failed paths can be shortened.

  In addition, when it is not possible to set up alternative paths for all paths that have failed using the pre-designed failure recovery procedure, the number of alternative path paths that can be set independently from each other from the start point to the end point of the working path Alternatively, a working path for setting an alternative path can be selected from a bandwidth that can be secured by these paths. Thereby, when a scramble occurs, a scramble can be avoided by resetting an alternative path of a path having a small number of independent routes to a different path. A path having a large number of independent paths has a high probability that an alternative path can be set for another path. Therefore, the recovery rate can be increased by performing the above.

  Further, in the dynamic design type failure recovery procedure, a link having a relatively small resource share rate can be preferentially selected. As a result, avoiding links with a high resource share rate for dynamic recovery, giving priority to selecting links with a low resource share rate, and setting an alternative path, thereby creating a probability of a situation in which other fault paths compete with spare resources. Can be avoided, and path setting failures during recovery can be reduced.

  In addition, a dynamic design type fault recovery procedure is performed after the pre-design type fault recovery procedure. In the dynamic design type fault recovery procedure, a pre-design that is set in advance for use in the working path and the pre-design type recovery is used. Sections other than the fault section of the path can be partially used. As a result, the range in which new route search and resource allocation are performed in the secondary recovery can be minimized, and the path setting process can be speeded up by the simplification of the setting.

  Further, when a resource shortage occurs in one link on the alternative path setting route in the failure recovery procedure, a resource allocation priority that is already allocated is low for a path with a high resource allocation priority. If a resource is diverted from one link on the alternative path setting route, the resource is assigned to another node on the alternative path route. The resource of the low-priority path in which the resource diversion occurred is transmitted regardless of whether or not there is a shortage of resources in all other links on the alternative path route. Can be included.

  In this way, resources of low-priority paths once resource diversion has occurred are preferentially input to failure recovery, so that resources of other useful paths can be preserved. In addition, when a failure occurs in another location, the recovery rate of failure recovery can be improved.

  A second aspect of the present invention is a management node that executes a procedure in the failure recovery method of the present invention, and stores an application scenario of a failure recovery procedure with a different path setting method or resource reservation method set for each path. A recovery scenario storage means; a recovery scenario management means for determining a recovery procedure with reference to a setting of the recovery scenario stored in the recovery scenario storage means when a failure occurs; and And means for transmitting to the device.

  A third aspect of the present invention is a communication node that executes the procedure in the failure recovery method of the present invention, the policy storage means for storing the policy setting for path setting, and the path setting method set for each path Or recovery scenario storage means for storing application scenarios of failure recovery procedures with different resource reservation methods, and when a failure occurs, the recovery scenario stored in the recovery scenario storage means is read, and route information and resource information of the entire network and the own node And a means for determining a specific path setting procedure and policy for each path with reference to the resource information and its blocking status.

  A fourth aspect of the present invention is a network composed of the management node of the present invention and the communication node of the present invention.

  A fifth aspect of the present invention is a program that, when installed in an information processing apparatus, causes the information processing apparatus to realize a function corresponding to the management node of the present invention or the communication node of the present invention.

  A sixth aspect of the present invention is the information processing apparatus-readable recording medium on which the program of the present invention is recorded. By recording the program of the present invention on the recording medium of the present invention, the information processing apparatus can install the program of the present invention using this recording medium. Alternatively, the program of the present invention can be directly installed in the information processing apparatus via a network from a server holding the program of the present invention. Thereby, the management node of this invention or the communication node of this invention is realizable using a general purpose information processing apparatus.

  According to the present invention, by applying a plurality of failure restorations in stages, it becomes possible to set up an alternative path of a failure path more flexibly than in the prior art, and the adaptability to failures, particularly the adaptability to multiple failures is improved. To do. In addition, by setting a path design policy in consideration of applying failure recovery in multiple stages to a node, the effect of improving the adaptability of the above method can be maximized.

  In order to show the effect of the invention, the following simulation is performed, and the result is shown in FIG. FIG. 4 is a diagram showing the concept of the physical network. FIG. 5 is a diagram showing the concept of a path network. FIG. 6 is a diagram showing the recovery rate improvement effect of the present invention. The working optical path networks shown in FIGS. 5A, 5B, and 5C are accommodated in the physical network shown in FIG. 4, and necessary resources are allocated thereto. Subsequently, an arbitrary single link failure is assumed, a spare optical path capable of recovering 100% of the failure is accommodated, and spare resources are allocated. Further, the ratio of the pre-design type restoration is 95% for the path network A, 92% for the path network B, and 84% for the path network C.

  Next, in the network designed as described above, a value R expressed by the following equation is defined as a recovery rate and calculated.

ｍ：数リンク数
Ｓ：全リンクからなる集合
ｐ（ｉ，ｊ）：リンクｉ，リンクｊの故障により影響を受ける現用光パス数
ｐ’（ｉ，ｊ）：ｐ（ｉ，ｊ）の内、残存リソースを利用することにより復旧可能な光パス数
Ｒを、（１）事前設計に従って復旧した場合、（２）事前設計により復旧し、続いて動的設計により復旧した場合のそれぞれパス網のケースについて求める。

m: number of links S: set of all links p (i, j): number of working optical paths affected by failure of link i and link j p ′ (i, j): of p (i, j) The number of optical paths R that can be recovered by using the remaining resources is as follows: (1) When recovered according to prior design, (2) Recovered according to prior design, and then recovered according to dynamic design. Ask for a case.

  As shown in FIG. 5, the results show that the present invention improves the recovery rate and improves the adaptability.

  In general, the recovery time of the pre-design type is on the order of several hundreds msec, and the recovery time of the dynamic design type is on the order of several seconds. In the present invention, in the double failure recovery, the ratio of the paths that can be recovered by the pre-design is calculated among the recoverable paths, and it is as follows, and many paths are recovered in the order of several hundred msec. It turns out that it is possible. When performing recovery using only dynamic design, it is necessary to recover all paths over a period of several seconds. Compared to this case, the present invention can significantly increase the speed. I can prove it.

  As described above, it is possible to dramatically improve the reliability of communication in the backbone network.

  A failure recovery method according to an embodiment of the present invention will be described with reference to FIGS. FIG. 7 is a diagram showing a configuration example of a network. As shown in FIG. 7, the embodiment of the present invention includes a plurality of communication nodes 1 to 7 that communicate with each other through connected links, and communication nodes 1 to 7. An example of a network composed of management nodes 8 that manage the network will be described. The failure recovery method according to the present embodiment is implemented by exchanging messages between the communication nodes 1 to 7 or between the communication nodes 1 to 7 and the management node 8.

  The management node 8 can request the communication nodes 1 to 7 to set the path calculated by the accommodation design engine through the management message, and the communication nodes 1 to 7 receive the message and set the path. It is also possible for the communication nodes 1 to 7 to issue a path setting request, calculate a route, and set a path. In the case of failure recovery, when a failure is detected in the communication nodes 1 to 7, the communication nodes 1 to 7 issue a working path deletion request and an alternative path setting request, and perform recovery operation in an autonomous and distributed manner. . Even when no failure has occurred, the management node 8 and the communication nodes 1 to 7 can perform the switching operation from the working path to the alternative path by issuing a request for switching to the alternative path.

  FIG. 8 is a diagram showing an outline of the management node. As shown in FIG. 8, the management node 8 includes an accommodation design engine 81 for setting a path in the network, a network management database 80 for managing network information, , A management message exchange link 82 for exchanging messages with the communication nodes 1 to 7, a recovery scenario management function unit 83, and a recovery scenario database 84. In the recovery scenario database 84, application scenarios of different failure recovery procedures can be set for each path, and the recovery scenario management function unit 83 determines the recovery procedure by referring to the settings of the recovery scenario database 84, and receives a management message. Can be used to communicate the application procedure and policy to the communication node.

  FIG. 9 shows an outline of the communication nodes 1 to 7. The communication nodes 1 to 7 include a switching hardware 11 and a switching hardware management control function unit 10. FIG. 10 shows the configuration of the switching hardware 11 of the communication nodes 1 to 7, and FIG. The configuration of the switching hardware management control function unit 10 is shown.

  The communication nodes 1 to 7 are connected to other nodes via a data communication link 12 that transmits / receives data to / from other nodes, and performs transmission / reception of communication data and path switching.

  The switching hardware management control function unit 10 is a device in which software for managing the switching hardware 11 is installed, and the switching hardware management control function unit 10 of another node via the control message link 20 for control message communication. Exchange control messages necessary for path setting.

  As shown in FIG. 10, the switching hardware 11 includes an optical switch unit (OPcon unit control unit) 30, an optical transmission unit (OPC unit control unit) 31, and an optical reception unit (OPC unit control unit) 32. Here, the optical switch unit 30 receives a signal transmitted from the adjacent node or output from the optical transmission unit 32, and outputs an optical signal to the adjacent node or the optical receiver unit 31. The interface (IF) has a function of detecting a path failure, and the detected failure information can be transmitted from the switching hardware 11 to the switching hardware management control function unit 10.

  As shown in FIG. 11, the signaling protocol processing unit 21 of the switching hardware management control function unit 10 is a function unit that processes a signaling protocol that performs path setting, switching, and deletion. It communicates with other nodes through the control message link 20. In addition, the session is managed for each path, and information regarding the set path and path signaling is held. The route calculation function unit 22 is a function unit that calculates a route for setting a path by transmitting a signaling message.

  The routing protocol processing unit 23 is a functional unit that processes a routing protocol, and collects network route information and resource information by exchanging messages with other nodes via the control message link 20.

  The switching hardware control middleware 24 is a functional unit that transfers control information such as a switch setting request based on a signaling protocol, a failure alarm from the switch, and the like between each functional unit and the switching hardware 11.

  The policy information DB (database) 25 is a database that holds policy settings for path setting, and is referred to by the route calculation function unit 22 in route calculation. The route information DB 26 is a database in which topology information and resource information of the entire network collected by the routing protocol are stored.

  The resource information database 27 is a database that stores the resource amount of the switching hardware 11 in its own node and its blocking status. RSVP-TE (for example, see Non-Patent Document 2), LDP (for example, see Non-Patent Document 3), etc. for the signaling protocol, OSPF-TE (for example, see Non-Patent Document 4), BGP4 (for example, for example) Non-Patent Document 5) and ISIS (for example, see Non-Patent Document 6) can be used.

  As the signaling in this embodiment, 2-way signaling such as RSVP-TE is applied. This is a form in which a Path message is transmitted from the start node of the path to the end node of the path, and the end node of the path returns a Reserve message to the start point. In the case of an active path or an alternative path setting without prior reservation, the availability of resources is confirmed with a Path message, and a switch is set with a Reserve message.

  The recovery scenario database 29 is a storage medium that stores application scenarios of different failure recovery procedures for each path, and can be set by a management message from a management node or directly set to a communication node. The recovery scenario management function unit 28 is a function unit that reads the recovery scenario set in the recovery scenario database 29, determines a specific path setting procedure and policy for each path, and transmits the information to the signaling protocol processing unit 21. is there.

  FIG. 12 shows an example of setting information set in the recovery scenario database 29. As shown in FIG. 12, in the recovery scenario database 29, firstly applicable types of failure recovery procedures are set. Second, a recovery scenario that combines the failure recovery procedures is defined. Third, a recovery scenario applied for each path ID is set.

  FIG. 13 shows an outline of path setting by signaling. When each communication node on the path relays the Path message, each communication node checks whether resources can be secured. If the resource cannot be secured, the transfer of the Path message is stopped and an Error message is returned. The Error message is returned to the beginning of the path. Further, the path is deleted by transferring the delete message to the communication nodes constituting the path.

  Further, when an alternative path is reserved in advance, switch setting is not executed when a path Reserve message is returned, and only resource reservation is performed. This is because the reserve resource is shared by putting the resource in the reserved state. In order to put the pre-reserved path into the operating state, it is necessary to perform signaling again on the reserved route after the failure occurs. However, since this signaling processing does not require route and resource calculation, it can perform processing faster than normal path setting.

  In this method, the path is set by transferring a signaling message created by the signaling protocol processing unit 21 on the route determined by the route calculation function unit 22 according to the policy set in the policy information DB 25.

  In this method, a pre-designed recovery method in which an alternative path is set in advance is applied to primary recovery, and a dynamic design recovery method in which an alternative path is immediately set after a failure occurs is applied to secondary recovery.

  The preset recovery path for primary recovery is set by the signaling process for the preset path simultaneously with the setting of the working path prior to the failure.

  The transition function of the failure recovery method is realized by cooperation of the signaling protocol processing unit 21 and the route calculation function unit 22. The failure information is detected by the IF of the optical switch unit 30 and notified to the switching hardware management control function unit 10.

  When a failure occurrence is notified to the switching hardware management control function unit 10, the signaling protocol processing unit 21 transmits a signaling message on the backup route until the failure information reaches the start point of the path. An alternate path is set by switching.

  The setting of the alternative path is performed by transferring the Path message and the Reserve message, similarly to the setting of the working path. When the error message is notified to the starting communication node that has started the alternative path setting signaling because the alternative path setting has failed, it is determined that the primary recovery has failed in the communication node in which the error message has occurred, and the secondary recovery is performed. Be started.

  The route calculation function unit 22 calculates a detour that does not pass through the failed link, and transmits an alternate path setting signaling message for secondary recovery to the route. The setting of the alternative path in which the switch is set is completed in the same manner as the working path.

  The outline in the case of performing the primary recovery and the secondary recovery is shown in FIGS. FIG. 14 is a diagram showing a route setting example of the working path and the pre-design backup path. FIG. 15 is a diagram showing an example of a setting sequence and a failure recovery sequence for the working path and the pre-design backup path. In FIG. 15, a Path message 1 and a Reserve message 1 represent a working path setting process (signaling). A Path message 2 and a Reserve message 2 represent reservation setting processing for a preset path. The Path message 3 is a setting process for changing to the operation state of the preset path in order to switch from the working path to the backup path when a link failure occurs between the communication nodes AB.

  If there is no failure, the message is transferred to the same route as the preset path. However, in the example of FIG. 15, this setting process is interrupted in the communication node C due to a link failure between the communication nodes CD. The error message 3 is transferred to the communication node A.

  The communication node A determines that the primary recovery has failed based on the information included in the error message, and additionally specifies the communication node and the failure link that issued the error message. Further, the fault link is removed from the path calculation topology, and a path that bypasses the fault link is calculated for secondary recovery. The transfer of the Path message 4 and the Reserve message 4 is a dynamic recovery process that is set according to a route calculated when the communication node A receives a failure.

Further, FIG. 16 shows an example of route setting when a method in which the entire path is used as the recovery section for primary recovery and a method in which one section on the path is used as the recovery section for secondary recovery is applied. Also in this case, the primary restoration and the secondary restoration are performed by the standard path setting procedure described above. For the calculation of the preset path used for primary recovery, route calculation is performed so that the entire working path is bypassed, and when performing secondary recovery route calculation, only the failed link is excluded from the route calculation topology, Route calculation that detours only.
a) When a plurality of failed paths are generated and an alternative path is set, if the alternative paths pass through the same link and resource contention occurs between the failed paths, the communication node at the starting point of the failed path The number q of independent routes from the communication node to the end communication node is compared, and recovery is performed with priority given to a fault path having a small q.

  The signaling procedure is as follows.

  When a faulty path occurs, as described above, fault information is notified to the communication node at the start point of the path, and the start-point communication node transmits a path message to start signaling for setting an alternative path. When transferring the Path message, each communication node confirms the availability of the resource of the link on the downstream side (end point node side) on the signaling path, and transfers the Reserve message downstream if it is available.

  The path setting signaling has a unique ID so that it can be distinguished from other signaling protocols, and the Path message includes the ID of the signaling session. The session ID and the starting communication node ID, By managing the communication node ID of the end point at each communication node, the signaling session is uniquely identified on the network.

  When a plurality of alternative paths are set in parallel, another alternative path signaling Path message arrives until a Reserve message is returned. When these multiple path messages are accepted, resources may be exhausted. If the Path message still arrives even if the resource is depleted, determine the priority of the alternative path setting including the alternative path that has already been transferred downstream, and the path with the lower priority will receive the resource. If it is used, the resource of the low priority path is assigned to the high priority path, and the setting of the low priority path is canceled.

  For a path to which the Path message has already been transferred downstream, when the Reserve message is returned, the Reserve message is changed to an error message and transferred upstream, and the failure of the alternative path is notified to the start point node.

  When performing resource reallocation from a low priority path to a high priority path, the path message that arrives after resource depletion that triggers resource reallocation is a low-priority path that has been canceled. The identification information such as the session ID and the Tunnel ID can be reused by bumping the resource of the low-priority path whose path setting has been canceled even in the link where the resource is not exhausted. The above procedure is shown in FIG. FIG. 17 is a diagram illustrating a sequence when resource reassignment to a path with a higher priority occurs during signaling. FIG. 18 is a configuration example of a network capable of executing the sequence of FIG.

  In addition, in the state where the free resource amount of each link of the entire network can be grasped by the routing protocol in each communication node, in addition to the method of selecting the set path during the path setting process as described above, the communication node Among the failed paths, there is a method of selecting a path for performing an alternative path setting process based on the priority order and performing the setting process only for the selected path.

  In the path selected in FIG. 18, the communication node A counts the number of paths passing between the communication node A and the communication node B for each setting process of the working path # 1, the working path # 2, and the working path # 3, and the number information thereof. Is stored in the own communication node, and at the same time, the free link information of the network is collected by the routing protocol and held in the own communication node.

  When a failure occurs in the link between the communication node A and the communication node B, the communication node A performs calculation based on the empty resource information that collects how many paths can be recovered in the empty resource range. The communication node A transmits a message including a determination result as to whether or not to switch to a switching end node (communication node that performs switching control, or an Ingress node in the case of path end switching) of each failed path, The switching end node reads the content of the received message and performs the switching process of the fault path only when it is determined that “switching is performed”.

  In FIG. 18, assuming that the restoration priority of the three paths affected by the failure between the communication node A and the communication node B is the working path # 1> the working path # 2> the working path # 3, the network in this case Since it is possible to determine that the number of recoverable resources is one based on the empty resource information, the communication node A determines that “switching” is performed with respect to the switching node (in this case, the communication node E) of the working path # 1. A message including the information determined to be “not switched” is transferred to the switching node (here, communication node A) of the working path # 2 A message including information determined to be “not to be switched” is transferred to the switching node of path 3 (here, communication node A). As a result, only the switching node of the working path # 1 including the information to be switched performs the alternative path setting process, and only the working path # 1 is restored.

  The number q of independent routes is managed as follows.

Prior to setting the working path, the route calculation function unit 22 of the communication node that starts signaling calculates the number q of independent routes from the path start point to the end point. When the signaling message is transferred, q is included in the message and transferred on the set path route. The communication node on the path manages the number of q in the signaling session. When failure recovery signaling is executed, information on q included in the Path message of failure recovery signaling is compared, and resources are preferentially allocated to signaling sessions with a small q.
b) When performing secondary recovery, a link with a low resource share is preferentially selected to set an alternative path.

When a pre-designed failure recovery method is applied to primary recovery, there is a backup path setting method in which resources are shared between different backup paths. In the present invention, the resource share rate of the link connected to the communication node in each communication node is calculated and held in the resource information DB 27. This resource share rate is exchanged between communication nodes by a routing protocol, and the resource share rate of all links in the entire network is held in each communication node. When selecting a route in the secondary recovery, the detour route is calculated by weighting the routing cost by the resource share rate.
c) When performing secondary recovery, make the best use of the pre-designed path with reserved resources.

  As described above, the notification of the error message at the time of the primary recovery signaling failure to the start node is a trigger for starting the secondary recovery. The route information of the pre-designed path is included in this error message, and the route information received from the error message is used in the route calculation of the alternative path set by the route calculation function unit of the source node in the secondary recovery. In the route calculation for secondary recovery, as the preset path, a route is selected so that the route for which resource reservation has been made in advance is included as much as possible, and the route for performing a new resource search is made as small as possible. Such a route selection method is shown in FIG. Secondary recovery signaling is performed in accordance with the route thus determined, and a path is set.

  That is, as shown in FIG. 19, a working path and a preset path (backup path) are set (S1), and when a failure of the working path is detected (S2), first, switching to a preset backup path is performed ( S3). At this time, if the backup path setting fails (S4), the path information of the failed working path and the path information of the preset path are acquired (S5). The route calculation function unit 22 obtains a route s1 independent of the working path and the preset path and sets the number of hops of the route to N (S6). A link end node that is in direct contact with the failed link is set as a start point and an end point, a route t that bypasses the failed link is obtained, and the hop count of t is set to M (S7). If N ≦ M (S8), s1 is adopted as a recovery path for secondary recovery (S9), an alternative path is set for the determined path (S11), and the recovery is terminated (S12). Here, if N> M (S8), a route s2 is obtained by replacing the portion corresponding to the failed link with t in the route of the working path, and s2 is selected as the recovery route for the secondary recovery (S10) and determined. An alternative path is set to the route (S11), and the recovery is terminated (S12).

  The present embodiment can be implemented as a program that causes an information processing apparatus to function as the above-described communication node or management node by being installed in a general-purpose information processing apparatus. This program can be recorded on a recording medium and installed in the information processing apparatus, or installed in the information processing apparatus via a communication line, so that the information processing apparatus can function as the above-described communication node or management node. .

  According to the present invention, it is possible to set up an alternative path for a fault path more flexibly than in the prior art, and adaptability to faults, particularly adaptability to multiple faults, is improved. Therefore, reliability for both network administrators and users is improved. In addition, convenience can be improved.

The flowchart which shows the procedure at the time of the start of operation | use of a path | pass. The flowchart which shows the procedure of a failure recovery process. The figure which shows the node independent path | route between Node-1 and Node-6 when Node-1 is made into a starting point, and Node-6 is made into an end point, and its resource. The figure which shows an example of a physical network. The figure which shows the example of a path | route network. The figure for demonstrating the recovery rate improvement effect by this invention. The figure which shows the structure of a network. The figure which shows the outline | summary of a management node. The figure which shows the outline | summary of a communication node. The figure which shows the outline | summary of switching hardware. The figure which shows the outline | summary of a switching hardware management control function part. The figure which shows an example of the setting information set to a recovery scenario database. The sequence diagram which shows the outline | summary of signaling. The figure which shows the example of a setting of an active path and a prior design backup path. The sequence diagram which shows the example of a setting of an active path and a prior design backup path. The figure which shows the example of the path | route set beforehand for the primary recovery, and the path | route of a secondary recovery. The sequence diagram which shows the outline | summary of the competition avoidance at the time of signaling. The network block diagram for demonstrating the outline | summary of the competition avoidance at the time of signaling. The flowchart which shows the procedure of a route selection method.

Explanation of symbols

1 to 7, A to F Communication node 8 Management node 10 Switching hardware management control function unit 11 Switching hardware 12 Data communication link 20 Control message link 21 Signaling protocol processing unit 22 Route calculation function unit 23 Routing protocol processing unit 24 Switching hardware control middleware 25 Policy information database 26 Path information database 27 Resource information database 28 Recovery scenario management function unit 29 Recovery scenario database 30 Optical switch unit 31 Optical transmission unit 32 Optical reception unit 80 Network management database 81 Accommodation design engine 82 Management message Replacement link 83 Recovery scenario management function unit 84 Recovery scenario database

Claims (15)

Consists of a plurality of communication nodes and a link connecting these communication nodes, and is applied to a network that performs communication by setting a path between any of the communication nodes, and when a failure occurs in the working path, In the failure recovery method in which the communication node sets an alternate path for the working path,
The communication node stores in advance a failure recovery scenario in which a plurality of failure recovery procedures having different path setting methods or resource reservation methods and the order of application of the plurality of failure recovery procedures are set. When an error occurs, the first failure recovery procedure is performed according to the failure recovery scenario. If an alternate path cannot be set by the failure recovery procedure, the failure that was first performed by the path setting method or resource reservation method A failure recovery method, wherein a failure recovery procedure different from the recovery procedure is performed next.   The message exchanged between the communication nodes in the failure recovery procedure includes information on whether or not to perform a failure recovery procedure different in path setting method or resource reservation method from the first failure recovery procedure. The failure recovery method according to 1.   In the message exchanged between the communication nodes in the failure recovery procedure, information indicating that a failure recovery procedure different in path setting method or resource reservation method from the first failure recovery procedure is to be executed next, and failure information, The failure recovery method according to claim 1, further comprising information relating to the recovery operation.   The plurality of failure recovery procedures include a pre-design type failure recovery procedure for performing alternative path route calculation before occurrence of a failure, and a dynamic design type failure recovery procedure for performing alternative path route calculation after a failure occurs. The failure recovery method according to 1.   The multiple failure recovery procedures include a failure recovery procedure in which the entire working path is used as a failure section and switched to an alternate path at both ends of the working path, and a part of the working path is used as a failure section and only the failure section is switched to another path. The failure recovery method according to claim 1, further comprising a failure recovery procedure as a path.   When a failure occurs on a link, the communication node or management node stores information on the number of active paths that will fail and the number of paths that can be set as an alternative path. 2. The working path for setting an alternative path is selected based on a policy set in advance in a communication node or a management node when the number of active paths is larger than the number of paths that can be set as an alternative path. Failure recovery method. When it is not possible to set an alternative path for all paths that have failed using the pre-designed failure recovery procedure,
The failure recovery method according to claim 4, wherein the working path for setting an alternative path is selected from the number of alternative path routes that can be set independently of each other from the start point to the end point of the active path or a bandwidth that can be secured by these routes.   5. The failure recovery method according to claim 4, wherein in the dynamic design type failure recovery procedure, a link having a relatively low resource share rate is preferentially selected. Perform dynamic design type failure recovery procedure after advance design type failure recovery procedure,
5. The failure recovery method according to claim 4, wherein in the dynamic design type failure recovery procedure, a section other than a failure section of a preset path set in advance for use in the active path and the predesigned recovery is partially used. . When a resource shortage occurs in one link on the alternative path setting route in the failure recovery procedure, a path with a low resource allocation priority already assigned to a path with a high resource allocation priority. Has a procedure to set the path by diverting resources
When resource diversion occurs on one link on an alternative path setting route, information on the path on which the resource diversion occurred is transmitted to other nodes on the alternative path route, and the other path on the alternative path route is transmitted. 2. The failure recovery method according to claim 1, further comprising a procedure of diverting a resource of a low priority path in which the resource diversion has occurred regardless of whether or not a resource shortage has occurred in all of the links. A management node for executing the failure recovery method according to claim 1,
A recovery scenario storage means for storing application scenarios of failure recovery procedures with different path setting methods or resource reservation methods set for each path;
When a failure occurs, a recovery scenario management means for determining a recovery procedure with reference to the setting of the recovery scenario stored in the recovery scenario storage means,
Means for transmitting the recovery procedure to the communication node using a management message. A communication node for executing the failure recovery method according to any one of claims 1 to 10,
Policy storage means for storing policy settings for path setting;
Recovery scenario storage means for storing application scenarios of failure recovery procedures with different path setting methods or resource reservation methods set for each path;
When a failure occurs, the recovery scenario stored in the recovery scenario storage means is read, and the route information and resource information of the entire network, the resource information of the own node, and the blocking status thereof are stored in the policy. Means for determining for each path based on policy settings held in the means.   A network comprising the management node according to claim 11 and the communication node according to claim 12.   A program that, when installed in an information processing apparatus, causes the information processing apparatus to realize a function corresponding to the management node according to claim 11 or the communication node according to claim 12.   The information processing apparatus-readable recording medium on which the program according to claim 14 is recorded.

Images