JP4686490B2 - Path switching method, path switching system, and branch node - Google Patents

Description

  The present invention relates to a technology for switching paths in a network that guarantees quality.

  Conventionally, in a network that guarantees quality, a method in which a backup path is provided redundantly in addition to a working path that is normally provided to a user is used in case a failure occurs in a path that is provided to a user. is there. FIG. 11A is a diagram for explaining a conventional switching method in a network having redundant paths (active path and backup path). In this case, in order to switch the active path to the standby path, the branch node that detects the failure switches to the standby path (Fast Reroute), or the branch node that receives the notification from the node that detects the failure is the standby path. Switching to a system path (Path Protection). In this method, the start point and end point (termination) of the backup path need to be included in the nodes constituting the active path. That is, the termination node of the protection path is the same as the termination node of the working path. For this reason, when a failure occurs at the terminal node of the active path, this redundant path switching method cannot cope.

  As a method for solving the above problem, in recent years, a network user's node is connected to a plurality of nodes of the network operator to make the communication path redundant, or conversely, for the network operator to make the communication path redundant. A method of connecting to a network user's node by a plurality of nodes is also used. FIG. 11B is a diagram for explaining a conventional method of transferring communication using IP routing. In this method, when a failure occurs in one of the nodes of a connected network operator, a packet is transmitted to a node at a different connection point by recalculating the route of IP (Internet Protocol) routing. Then, transfer to a node where no failure has occurred and resume communication. As a result, even when a failure occurs in the terminal node of the working path, it is possible to cope with it.

  Here, when the switching method in the redundant path shown in FIG. 11A is compared with the method of switching communication by route calculation of IP routing shown in FIG. 11B, the former is faster, but the termination node It is not possible to cope with the failure of. On the other hand, in the latter case, it is possible to cope with a failure of the terminal node, but it is necessary to calculate a route for IP routing, and it takes time to recover the communication. Furthermore, in the latter case, since communication is temporarily interrupted, more communication packets are lost.

  A multicast switching method has been proposed as a method that can cope with a failure of a terminal node and minimize the loss of communication packets (see Non-Patent Document 1). FIG.11 (c) is a figure explaining the switching method by the conventional multicast. There are two switching methods by multicast.

(1) A node (branch node) that is a branch point of the working path and the protection path transmits a communication packet to both the working path and the protection path. Then, the terminal node of the backup path discards the received communication packet in order to prevent double transmission. The branch node always monitors that all nodes and links in the working path are functioning normally using a health check, etc., and if a failure is confirmed, discards it at the end node of the backup path. Notification to cancel.

(2) A node (branch node) that is a branch point of the active path and the standby path transmits a communication packet only to the active path when it is normal, and transmits a communication packet only to the standby path when a failure occurs. The branch node constantly monitors that all nodes and links of the working path are functioning normally by a health check or the like, and switches the transmission to the backup path when the occurrence of a failure is confirmed.
Shimizu Shigeki, Hiroyuki Ouchi, “NW Redundancy with Multicast MPLS”, IEICE, “2005 Society Conference Proceedings”

  However, in the switching method using multicast, a multicast function that is a complicated mechanism is required at a branch node that is a branch point between the active path and the backup path, and thus the processing load on the branch node is increased. In addition, the branch node must always monitor the downstream node of the active path, and the load is applied to the network in addition to the load of the branch node by the monitoring process.

  Accordingly, an object of the present invention is to solve the above-described problems, to cope with a failure of a terminal node of an active path, to realize high-speed failure recovery, and to reduce the load on the node and the network.

In order to solve the problem, the path switching method according to claim 1 is a path switching method for switching from a working path to a backup path having a different termination node with respect to the working path in a network, When a node that becomes a branch point between the working path and the backup path includes a storage unit that stores routing information in the network, and obtains route information of the working path and the backup path, the routing Referring to the information, the information on the interface for sending the communication packet to the working path and the information on the interface for sending the communication packet to the protection path, the working path and the protection path and storing in the storage unit in association with the use of the same path ID, the working path is available in the Detecting link down of the nodes of the link, or the from the downstream node working path to receive the failure information including the path ID, the step of detecting whether a failure has occurred in the working path via which interface, The backup path with the same path ID as the path ID corresponding to the interface where the failure is detected is extracted with reference to the storage unit, and a communication packet is transmitted using the interface of the extracted backup path And a step of performing.

According to a third aspect of the present invention, there is provided a path switching system including a path including a working path and a backup path having a different termination node with respect to the working path in a network including a plurality of nodes. In the switching system, the node includes a storage unit that stores routing information in the network, a path ID for identifying a path in the network, a start node of the working path, and a terminal node of the working path A working path setting request including at least information on path information of the working path, a terminal node of a protection path corresponding to the working path, information on the working path and a branch node that branches the protection path. Based on the acquired working path setting request, when it is determined that the own node is the starting node of the working path, Path setting means for transmitting a working path setting request to a downstream node of the working path based on route information of the working path, and based on the acquired working path setting request, the own node When it is determined that the node is a branch node, the path of the protection path connecting the start path node of the active path to the end node of the protection path included in the active path setting request is determined as the end of the active path. A path ID that is the same as the path ID of the working path, the start node of the working path , and the backup system , calculated so as not to overlap with a path downstream from the own node indicated in the path information of the node and the working path A path termination node, branch processing means for creating a backup path setting request including at least path information of the backup path, and the active path setting request as a path of the active path Path setting means for transmitting to the downstream node of the working path based on the information, and transmitting the created backup path setting request to the downstream node of the protection path based on the path information of the protection path. , Referring to the routing information, information on an interface for sending a communication packet to the working path, and information on an interface for sending a communication packet to the protection path, the working path and the Storage means for storing the same path ID in association with the backup path, detection of link down of the link of the own node used by the active path, or from a downstream node of the active path When a failure in the working path is detected by receiving failure information including the path ID, a failure occurs in the working path through which interface. Failure to send a communication packet using the interface of the backup path with the same path ID as the path ID corresponding to the interface in which the failure has occurred is determined with reference to the storage means Detecting means, and when the own node is not one of the start node of the active path, the end node of the active path, the end node of the backup path, and the branch node based on the active path setting request Storage means for storing at least the path information of the working path included in the working path setting request, and the working path setting request based on the path information of the working path. Refer to path setting means for transmitting to the downstream node and route information of the working path stored in the storage means when a failure of the working path is detected. Upstream node of the generated working path failure comprises failure detection means for transmitting the failure information including the path ID, based on the working path setting request, the own node is a terminal node of the working path A storage means for storing at least route information of the active path included in the active path setting request, and the active path stored in the storage means when a failure of the active path is detected. It comprises failure detection means for referring to the path information of the system path and transmitting the failure information including the path ID to the upstream node of the active system path where the failure has occurred.

The branch node according to claim 5 is a branch node that switches to a backup path corresponding to an active path in the event of a failure in a network composed of a plurality of nodes, and stores routing information in the network. A path ID for identifying a path in the network received from another node, a start node of the active path, an end node of the active path, route information of the active path, Based on the active path setting request including at least information on the end node of the standby path corresponding to the active path, the active path, and the branch node that branches the backup path, the start path node from the start node of the active path The path of the backup path that connects the end node of the backup path, the path information of the end node of the active path and the path of the active path Shown, calculated so as not to overlap with the downstream path from the local node, the same path ID and path ID of the working path, the start node of the working path, an end node of the backup path, of the backup path A branch processing unit for creating a standby path setting request including at least path information; and transmitting the working path setting request to a downstream node of the working path based on the path information of the working path, A path setting unit that transmits the protection path setting request made to the downstream node of the protection path based on the path information of the protection path, and sends a communication packet to the working path with reference to the routing information. and interface information for, and interface information for sending a communication packet to the backup path, wherein the same path in the working path and the backup path Fault information including the storage unit in association with the use of the ID, detection of the link down link of the node that the working path is using, or the path ID from the downstream node of the working path When the failure of the working path is detected, it is determined by referring to the storage unit which interface has failed through the interface, and the determined failure has occurred. And a failure detection unit that transmits a communication packet using the interface of the backup path with the same path ID as the path ID corresponding to the interface.

As a result, the path setting node transmits a path setting request including the setting information of the protection path, and the branch node collects not only the active path but also information regarding the protection path based on the received path setting request. Manage. Since the branch node can transmit and receive packets based on this information, flexible and high-speed failure recovery is realized, and the processing load on the node is reduced.
Further, according to this configuration, when a failure is detected in the relay node, the failure node is notified to the upstream node, whereby the branch node is switched to the backup path. For this reason, the number of backup paths to be set can be reduced, and it is possible to cope with failure detection of all nodes.

  According to a second aspect of the present invention, in the path switching method according to the first aspect, the working path and the backup path use MPLS LSP.

  According to this method, it is possible to realize quality assurance by MPLS together with path control.

According to a fourth aspect of the present invention, in the path switching system according to the third aspect, the information included in the acquired working path setting request includes a plurality of end nodes of the protection path for the working path, When the node determines that the own node is the branch node based on the acquired active path setting request, the branch processing means includes the end information of the active path and the path information of the active path. The route to the plurality of end nodes of the acquired backup path is calculated so that it does not overlap with the route downstream from its own node, and at least one of cost and delay among the calculated paths of the backup path route backup path but as a minimum, end nodes and select the route information, the same path ID and path ID of the working path, the start node of the working path End node of the backup path, characterized by creating at least backup path setting request including the routing information of the backup path.

According to a sixth aspect of the present invention, in the branching node according to the fifth aspect , the branch processing unit includes a plurality of backup paths for the working path that are included in the received working path setting request. The branch processing unit of the branch node includes downstream end nodes of the active path and the downstream of the own node indicated in the path information of the active path included in the active path setting request. Routes to the plurality of end nodes of the obtained backup path are calculated so as not to overlap with the paths, and among the calculated paths of the backup path, a path of the backup path that minimizes at least one of cost and delay is calculated. route, by selecting the termination node and route information thereof, the same path ID and path ID of working path, the start node of the working path, an end node of the backup path, the preliminary Characterized by creating a path at least backup path setting request including the routing information.

  As a result, when the branch node performs route calculation of the backup path, it is possible to select a route having a terminal node that is the shortest route from a plurality of route candidates for the backup path. The efficiency can be further improved.

  According to the present invention, it is possible to cope with a failure of a terminal node of an active path, realize high-speed failure recovery, and reduce the load on the node and the network.

  Hereinafter, the best mode for carrying out the present invention (hereinafter referred to as “embodiment”) will be described. Note that the present embodiment will be described on the assumption that when a path is switched in a network that guarantees quality, the path is switched to a path having a different termination point.

FIG. 1 is a diagram illustrating a network configuration example of a communication system to which the path switching system of the present embodiment is applied.
The communication system includes a core network 3 that is a network that performs quality assurance, and one or more user networks 2 (2a, 2b) that are operated by a general user, and are connected to be communicable with each other. The user network 2 and the core network 3 include nodes as communication devices using IP, and communication using IP is possible between nodes. Note that these networks (reference numerals 2 and 3) may be configured by nodes that can communicate with IPv6 or the like.

  The core network 3 is a network configured with a wideband link and performing quality assurance by using MPLS (Multi-Protocol Label Switching) or the like. The core network 3 includes a node 1 such as an MPLS router that connects to the user network 2. Among these nodes 1, the node 1 that is located at the boundary with another network and directly communicates with a node of another adjacent network is particularly called a provider edge.

  The user network 2 is configured by a LAN (Local Area Network) such as Ethernet (registered trademark), for example, and nodes such as a network hub and a router are connected by a LAN network cable. Among these nodes, a node that is located at a boundary with another network and directly communicates with another adjacent network is particularly called a user edge. The user edge is communicably connected to a plurality of provider edges of the core network 3, and these nodes can perform IP communication with each other.

  In this embodiment, when a communication packet is transmitted from a user edge belonging to the user network 2a to a user edge belonging to the user network 2b, the provider edge connected to be communicable with the user edge of the user network 2a is connected to the user network 2b. An active path is set between the user edge and the provider edge that is communicably connected. Then, as a redundant configuration, a standby path that branches from a branch node in the working path and that has a provider edge that is different from the provider edge that is the termination point of the working path is prepared for when a failure occurs. .

  Note that the connection destination of the provider edge of the core network 3 is not limited to the user network 2 and may be another core network 3. In this embodiment, the paths set by the provider edge (active path and backup path) are assumed to be connected between provider edges in the core network 3, but may extend over a plurality of core networks 3. The user network 2 may be included.

<Node configuration>
FIG. 2 is a block diagram showing a configuration of nodes belonging to the core network of FIG. Although not shown, the node 1 is configured as a computer including at least an input / output interface that controls input / output, a memory serving as a storage unit used when performing arithmetic processing, and an arithmetic processing device that performs the arithmetic processing. .

  As shown in FIG. 2, the node 1 includes a path setting unit 21 that sets a path, a label table 22 that is information for transmitting a packet, a branch processing unit 23 that determines the role of the node in the path, and a failure. A failure detection unit 24 to detect, a path information table 25 in which path information of a path is stored, and the like are configured. The functions of the path setting unit 21, the branch processing unit 23, and the failure detection unit 24 are realized by executing a program stored in a memory by a CPU (Central Processing Unit) as an arithmetic processing unit in the node 1. . The label table 22 and the path information table 25 are stored in a RAM (Random Access Memory), a flash memory, an HDD (Hard Disk Drive), or the like in the node 1.

  The path setting unit 21 receives a path setting request, which is a request to set a path, from a CLI (Command Line Interface) from a network operator, a device (management device) that manages the entire network, another node, and the like. Set. In addition, a path setting request is transmitted to another node 1, and information included in the path setting request acquired from the other node 1 is stored in the label table 22 and the path information table 25. Table 1 below shows information included in the path setting request.

  As shown in Table 1, the path setting request includes path setting information and backup path setting information. The path setting information passes between the path ID for identifying the path in the network, the starting node of the path to be set, the terminal node (provider edge) of the path to be set, and the path from the starting node to the terminal node of the path to be set. Information indicating a node or the like. The backup path setting information includes the start node of the path to be set, the branch node that branches the active path and the standby path of the path to be set, the end node (provider edge) of the active path, and the end of the backup path Information indicating a node (provider edge) or the like. Note that the information of each node is described by, for example, an IP address or a node number.

  Note that the path setting request includes label information of communication packets transmitted and received by the node 1, but illustration thereof is omitted here.

  The label table 22 is information that defines information for changing the label of the communication packet received by the node 1 and from which IF (interface) the communication packet is transmitted.

  In the present embodiment, in order to realize quality assurance in the core network 3, a transmission method using a label by MPLS LSP (Label Switched Path) or the like is used. The label is information used as a key for communication packet transmission, and information set based on a predetermined rule is added to the communication packet. There is also a method of using a value included in a communication packet in advance, a wavelength used for the communication, or the like as a label.

  When a communication packet is transmitted using a label by MPLS or the like, the first node of the path that receives the communication packet first adds a label to the received communication packet and is connected downstream (that is, connected to its own node). The node is transmitted to a node near the terminal node in the path. The downstream node that has received the communication packet acquires label information, replaces the label based on the information in the label table 22 stored therein, and transmits the communication packet from the corresponding transmission IF. An example of the label table 22 is shown in Table 2 below.

  Table 2 is a table showing an example of the label table 22 stored in the branch node in the present embodiment, and information on both the working path and the backup path is described with the same path ID. In other words, the same path ID of the working path and the path ID of the protection path corresponding to the working path are stored, so that the branch node has a correspondence relationship between the working path and the protection path. Can be determined.

  The label table 22 is information indicating a path ID for identifying a set path, an attribute, a reception label, a transmission label, a transmission IF, and the like. For the path ID, information on the path ID included in the path setting request is acquired and stored. The attribute is information for determining which record has priority when there are a plurality of settings (records) of the same path ID in the label table 22. The reception label is information for determining the label added to the received communication packet, and the label information of the communication packet included in the path setting request is acquired and stored. The transmission label is information for changing the label added to the received communication packet. Each time a record is registered, this delivery label is stored with a unique number in the delivery label of the label table 22. The sending IF is an IF for sending a communication packet. Although not shown here, routing information stored in a memory in the node 1 (information storing link information and topology information for connecting each node existing in the network) ), Information on the interface connected to the downstream node is acquired and stored. The transmission and reception of communication packets using the label and label table 22 will be described in detail with reference to FIG.

  The branch processing unit 23 determines which role the node itself is based on the information included in the received path setting request.

  The failure detection unit 24 detects a failure in the transmission IF used for packet transmission, a link down of the link of the own node used by the active path, or receives failure information from the downstream node If the path itself is a branch node, the information in the label table 22 is switched from the active path to the backup path. If the path is not a branch node, refer to the “route” in the path information table 25 described later. Then, the failure information is transmitted to the upstream node of the path where the failure has occurred (that is, the node connected to the own node that is close to the starting point node in the path).

  The path information table 25 is information indicating a path for each path, and is referred to when the failure detection unit 24 transmits the failure information to the upstream node. The path information table 25 is stored by the path setting unit 21 based on information such as the received path setting request. An example of the path information table 25 is shown in Table 3 below.

  Table 3 is a table showing an example of the path information table 25 stored in the branch node in the present embodiment. The path information table 25 is information indicating a path ID for identifying a set path, a start node of the set path, a terminal node of the set path, a route of the set path, and the like. The information included in the path information table 25 is stored by acquiring the information included in the path setting request received by the path setting unit 21. With regard to “route”, the path calculated by itself or the downstream node is stored. Alternatively, a route obtained from the route information included in the path setting confirmation information received from may be stored.

  The “path ID” in the path information table 25 and the “path ID” in the label table 22 are linked.

  FIG. 3 is a diagram conceptually illustrating a process in which each node transmits a communication packet using a label in the core network illustrated in FIG. 1. Hereinafter, for the sake of easy understanding, the node 1 is illustrated with an alphabetic symbol assigned to each node and described as appropriate. In FIG. 3, the setting of paths (active path and backup path) has already been completed, and the label table 22 and path information table 25 are stored in each node 1.

  In the network shown in FIG. 3, first, when node A, which is the starting node, receives a communication packet, the communication packet is labeled “5” based on the transmission label “5” of the label table 22 (reference numeral 301) stored in itself. And a communication packet is transmitted from IF “1” based on transmission IF “1”. It should be noted that the receiving label of the label table 22 (reference numeral 301) of the node A describes information indicating the starting point.

  When the node B, which is a branch node, receives the communication packet with the label “5”, it normally uses the working path. Therefore, based on the label table 22 (reference numeral 311), the label “5” of the communication packet is used. Is replaced with “3”, and a communication packet is transmitted from IF “1”.

  When the node F, which is the relay node of the working path, receives the communication packet with the label “3” added, the node F assigns the label “3” of the communication packet based on the transmission label “2” of the label table 22 (reference numeral 321). In place of “2”, the communication packet is transmitted from IF “1”.

  In the label table 22 (reference numeral 331) of the node H that is the active system end node, information indicating the end is described in the transmission label. When the node H receives the communication packet to which the label “2” is added, the node H removes the label based on the information indicating the end of the transmission label in the label table 22 (reference numeral 331), and performs normal packet transmission processing. Do.

  Note that the node B which is a branch node, when a failure occurs in the active path (for example, when a failure occurrence is detected by IF “1” of the node B), the attribute “standby system” in the label table 22 (reference numeral 311). Is switched based on the information “”, the label “5” of the received communication packet is changed to “4”, and the communication packet is transmitted from the transmission IF “2”. A detailed description of this switching process will be given later.

<Node behavior when setting a path>
The operation of the node 1 when setting a path will be described with reference to FIGS. FIG. 4 is a flowchart showing the operation of the node when setting a path.

<Start node (path setting node)>
The path setting unit 21 of the node 1 receives a path setting request from a control from the network operator via the CLI, a management device that manages the entire network, or the like (S101). The information included in this path setting request is as illustrated in Table 1 above.

  The branch processing unit 23 of the node 1 compares the information included in the received path setting request with its own identification information (for example, an IP address) to determine the role of the own node in the path to be set. (S102).

  In the determination in step S102, when the identification information of itself and the information of the “starting node” included in the path setting request match (“starting node” in step S102), the path setting unit 21 of the node 1 has received The path setting information (start node, end node, via) included in the path setting request is used to calculate the path of the active path (S111), and the information of the node that passes through based on the calculated path of the active path Is stored in “Via” of the path setting information received in step S101. In the process of step S111, when the information is not stored in the “path” of the path setting information included in the path setting request, or the path calculation is performed in the “path” of the path setting information included in the path setting request. It is executed when it is specified.

  Then, the path setting unit 21 transmits an active path setting request to the downstream node of the active path based on the “route” information included in the path setting information and the routing information stored in itself (see FIG. S112). If the path setting information included in the received path setting request contains all the information from the start node to the end node in advance, the route calculation processing in step S111 is unnecessary, and the path setting unit 21 The path setting request received in step S101 is transmitted as a working path setting request to the downstream node of the working path.

  When the path setting confirmation indicating that the path setting is successful is received from the downstream node (S113), the path setting unit 21 of the node 1 receives the path setting request received in step S101 and the path setting confirmation received in step S113. The information acquired from the above is stored in the label table 22 and the path information table 25 based on the routing information stored in itself (S114). Here, information exemplified by reference numerals 301 and 302 in FIG. 3 is stored. That is, the path ID information included in the path setting request is stored in the path IDs of the label table 22 and the path information table 25, and the path of the path information table 25 (reference numeral 302) is the path calculated in step S111. Alternatively, route information included in the path setting confirmation received from the downstream node is stored.

<Branch node>
On the other hand, the path setting unit 21 of the node 1 receives a path setting request from another node (upstream node) (S101), and is included in the identification information and the path setting request by the branch processing unit 23 in step S102. When the “branch node” information matches (“branch node” in step S102), the path setting unit 21 of the node 1 sends the “route” information included in the path setting information and the routing stored in itself. Based on the information, the path setting request received in step S101 is transmitted (transferred) to the downstream node as a working path setting request (S121).

  When the path setting confirmation is received from the downstream node (S122), the path setting unit 21 of the node 1 instructs the branch processing unit 23 to perform route calculation of the backup path, and the branch processing unit 23 Is calculated (S123).

  More specifically, the path setting unit 21 of the node 1 transmits the path setting request received in step S101 to the branch processing unit 23 as a backup path path calculation instruction when the path setting confirmation is received in step S122. To do. At this time, when the path information of the working path is added to the path setting confirmation received in step S122, the path setting request is transmitted to the branch processing unit 23 together with the path setting request. Based on the received path setting request (if added to the path setting confirmation, the branch processing unit 23 also calculates the path of the protection path). In the backup path path calculation, the branch processing unit 23 calculates the backup path path from the start node to the backup path end node so that it does not overlap with the active path path. In addition, when multiple end nodes of the backup path are specified (included), route calculation is performed for all end nodes, and the link cost and delay of the calculated route are the highest. Choose a route that will be smaller. The branch processing unit 23 transmits a backup path setting request storing the calculated backup path path to the path setting unit 21. Table 4 below shows information included in the standby path setting request transmitted from the branch processing unit 23 to the path setting unit 21.

  As shown in Table 4, in the protection path setting request transmitted from the branch processing unit 23 to the path setting unit 21, the termination node of the protection path is stored in the termination node of the path setting information. The route based on the backup path information calculated in step S123 is stored. The path setting information may include information indicating a standby system.

  The path setting unit 21 receives the backup path setting request received from the branch processing unit 23 based on the “route” information included in the path setting information and the routing information stored in itself. Transmit to the downstream node (S124).

  Subsequently, the path setting unit 21 uses the label table based on the routing information stored in the path setting request received in step S101 and step S124 and the information acquired from the path setting confirmation received in step S122. 22 and the path information table 25 (S125). The process of step S125 is the same as the process of step S114. Here, information exemplified by reference numerals 311 and 312 in FIG. 3 is stored.

  Then, the path setting unit 21 confirms the path setting confirmation indicating that the path has been set in the “route” information included in the information of the path setting request received in step S101 and the routing information stored in itself. Based on this, the data is transmitted to the upstream node (S126). At this time, downstream path information may be included in the path setting confirmation.

<Relay node>
On the other hand, the path setting unit 21 of the node 1 receives the path setting request from the other node 1 (upstream node) (S101), and the branch processing unit 23 determines the identification information and the path setting request in step S102. If the information does not match any of the included “start node”, “end node”, and “branch node” (“relay node” in step S102), the path setting unit 21 of the node 1 is included in the path setting information. Based on the “passing” information and the routing information stored in itself, the path setting request received in step S101 is transmitted (transferred) to the downstream node as a working path setting request (S131).

  When the path setting confirmation is received from the downstream node (S132), the path setting unit 21 of the node 1 uses the information acquired from the path setting request received in step S101 and the path setting confirmation received in step S132. Is stored in the label table 22 and the path information table 25 based on the routing information stored in (S133). Here, information exemplified by reference numerals 321 and 322 in FIG. 3 is stored.

  Subsequently, the path setting unit 21 performs path setting confirmation indicating that the path has been set, “route” information included in the path setting request information received in step S101, routing information stored in itself, and Is transmitted to the upstream node (S134). At this time, downstream path information may be included in the path setting confirmation.

<Current terminal node>
On the other hand, the path setting unit 21 of the node 1 receives a path setting request from the other node 1 (upstream node) (S101), and the branch processing unit 23 determines that its own identification information is included in the path setting request. When it matches the information of the “active terminal node” included (“active terminal node” in step S102), the path setting unit 21 of the node 1 uses the information acquired from the path setting request received in step S101. Is stored in the label table 22 and the path information table 25 based on the routing information stored in (S141). Here, information exemplified by reference numerals 331 and 332 in FIG. 3 is stored.

  Subsequently, the path setting unit 21 performs path setting confirmation indicating that the path has been set, “route” information included in the path setting request information received in step S101, routing information stored in itself, and Is transmitted to the upstream node based on the above (S142). At this time, downstream path information may be included in the path setting confirmation.

<Standby node>
On the other hand, the path setting unit 21 of the node 1 receives a path setting request from the other node 1 (upstream node) (S101), and the branch processing unit 23 determines that its own identification information is included in the path setting request. When it matches the information of the “standby end node” included (“standby end node” in step S102), the path setting unit 21 of the node 1 uses the information acquired from the path setting request received in step S101. Is stored in the label table 22 and the path information table 25 based on the routing information stored in (S151). For the backup path, a path setting confirmation may be sent as in the active path.

  Note that the timing at which the node 1 stores information in the label table 22 and the path information table 25 is not limited to the above description. For example, after transmitting the path setting confirmation, the information may be stored in the label table 22 and the path information table 25, or may be stored immediately after the node determination processing in step S102.

  In addition, “pass” of the path setting information included in the path setting request may not be specified (not included) in advance. In that case, not only the start node but also a branch node or a relay node, the route calculation as in step S111 is performed, and the calculated route information is stored in the path setting request and sent to the downstream node. .

<Flow of processing when setting a path>
FIG. 5 is a sequence diagram showing the flow of processing of each node in the network shown in FIG. Hereinafter, for description, a node (node A in FIG. 3) that plays a role of setting a path is referred to as a path setting node. A node (node B in FIG. 3) that plays a role of branching between the active path and the backup path is called a branch node. Further, the node (node H in FIG. 3) that is the termination of the working path is called the working termination node, and the node that is the termination of the protection path (node I in FIG. 3) is the protection termination node. Call it. Nodes that play a role of relaying communication packets (nodes C, E, and F in FIG. 3) are called relay nodes. The operation of each node (A, B, F, H, I) in FIG. 5 will be described with reference to FIG.

  First, the path setting node (node A) receives a path setting request from a CLI from a network operator, a management device that manages the entire network, or the like (S511). The information included in this path setting request is as illustrated in Table 1 above.

  When determining that the “starting node” included in the received path setting request matches the identification information of the path setting node (S512), the path setting node calculates the path of the active path (S513), and calculates the calculated active system. Based on the path of the path, information on the node through which the path is routed is stored in the path setting information of the path setting request. As described above, when the route information is stored in advance in the path setting information included in the received path setting request, the route calculation process in step S513 is not necessary.

  Then, the path setting unit 21 transmits an active path setting request to a branch node (node B) that is a downstream node of the active path (S514).

  The branch node (node B) receives the active path setup request from the path setup node (S521), and determines that the “branch node” included in the received path setup request matches its own identification information. (S522) The active path setting request is transmitted to the relay node (node F), which is the downstream node of the active path (S523).

  The relay node (node F) receives the active path setting request from the branch node (S531), and its own identification information includes the “start node”, “end node”, “branch” included in the received path setting request. If it is determined that it does not match any information of the node (S532), a working path setting request is transmitted to the working terminal node (node H), which is a downstream node of the working path (S533).

  The working end node (node H) receives the working path setting request from the relay node (S541), and the “working end node” included in the received path setting request matches the identification information of itself. (S542), information included in the path setting request received in step S541 is stored in the label table 22 and the path information table 25 (S543). Then, a path setting confirmation indicating that the path has been set is transmitted to the relay node (node F) that is the upstream node (S544). At this time, route information may be included in the path setting confirmation and transmitted.

  When the relay node (node F) receives the path setting confirmation from the working terminal node (S534), the relay node (node F) receives the information included in the path setting request received in step S531 and the path setting confirmation received in step 534, and the label table 22 and It stores in the path information table 25 (S535). Then, the path setting confirmation is transmitted to the branch node (node B) which is the upstream node (S536). At this time, route information may be included in the path setting confirmation and transmitted.

  When the branch node (node B) receives the path setting confirmation from the relay node (S524), it calculates the path of the backup path (S525), and sends the backup path setting request to the downstream node of the backup path. Transmit to the relay node (node C) (S526). The route calculation and transmission processing of the backup path here is as described in steps S123 and S124 of FIG.

  Then, the branch node stores the information included in the path setting request received in step S521 and the path setting confirmation received in step S524 in the label table 22 and the path information table 25 (S527), and performs the path setting confirmation upstream. The data is transmitted to the path setting node (node A), which is a node (S528).

  When the path setting node (node A) receives the path setting confirmation from the branch node (S515), the path setting request received in step S511 and the information included in the path setting request received in step S515 are displayed in the label table 22 and the path. It stores in the setting table 25 (S516).

  In step S526, the standby path setting request transmitted by the branch node is received by the relay node (node C), which is the downstream node of the standby system (not shown). The relay node stores information in its own label table 22 and path information table 25, and transmits the received standby path setting request to the downstream node.

  The standby terminal node (node I) receives the standby path setting request transmitted by the relay node (node E) (S551), and the “backup terminal node” included in the received path setting request is If it is determined that it matches the identification information (S552), the information included in the path setting request received in step S551 is stored in the label table 22 and the path information table 25 (S553).

  With the above processing, in the network shown in FIG. 3, the working path (node A-node B-node F-node H) and the backup path branched from node B (node A-node B-node E-node I) ) And a label table 22 and a path information table 25 are stored in each node 1. As described above, the timing at which the node 1 stores information in the label table 22 and the path information table 25 is not limited to the above description. For example, it may be stored immediately after the node determination process, or information may be stored in the label table 22 and the path information table 25 after transmitting a path setting confirmation. Further, information other than the “route” in the path information table 25 is stored in advance, and after receiving the path setting confirmation, the “route” in the path information table 25 is based on the path information included in the path setting confirmation. It is good also as storing information in.

  A path setting request transmitted by a path setting node (starting node) adds a different path ID for each path setting request. This makes it possible to manage paths with a common ID in each node. Hereinafter, a specific example will be described.

  FIG. 6 is a diagram showing a modification of the network shown in FIG. In the network configuration shown in FIG. 6, there are two start point nodes (node A and node X), and each start point node transmits a path setting request, so that two paths are set. More specifically, when the path setting request (1) is transmitted from the node A, the working path (node A-node B-node F-node H) having the path ID "1" and the path ID "1" are transmitted. A backup path (node A-node B-node C-node E-node I) is set. On the other hand, when the path setting request (2) is transmitted from the node X, the working path (node A-node B-node Y-node Z) with the path ID "2" and the standby system with the path ID "2" A path (node X-node B-node Y-node W) is set.

  As a result, the label table 22 (reference numeral 401) and the path information table 25 (reference numeral 402) are stored in the node B. In addition, a label table 22 (reference numeral 411) and a path information table 25 (reference numeral 412) are stored in the node Y. In this way, in the label table 22 and the path information table 25, the path is managed using a common ID for the active path and the backup path. For example, even when a working path is set from one start node to different end nodes, a different ID is added for each path setting request.

<Node behavior in the event of failure>
Next, the operation of the node 1 at the time of failure will be described with reference to FIGS. FIG. 7 is a flowchart showing the operation of the node at the time of failure. FIG. 8 is a diagram conceptually showing the operation of each node at the time of failure in the network shown in FIG. Note that, as shown in FIG. 8, the label table 22 and the path information table 25 are stored in advance in the node B and the node F in FIG.

  The node 1 detects a failure by the failure detection unit 24 (S601). As a method for detecting a failure, for example, there is a method of detecting a failure of an IF used for packet transmission, detecting a link down of a link of the own node, or receiving failure information from another node 1.

  The failure detection unit 24 refers to the label table 22 stored in itself, and determines whether or not there is a switchable backup path with the path ID corresponding to the IF that detected the failure in step S601 (S602). .

  When there is no switchable backup path with the path ID corresponding to the IF that detected the failure (S602 → No), the failure detection unit 24 of the node 1 has the path ID stored in the path information table 25. Referring to the route, based on the routing information stored in itself, the failure information including the path ID information is transmitted to the upstream node of the own node (S603).

  In FIG. 8, a case where the link between the node F and the node H is down is shown as an example. This example will be specifically described. First, the node F detects a failure at IF “1”. The node F refers to the label table 22 (reference numeral 811), and the IF “1” in which the failure is detected is the attribute “active” of the path ID “1”, and the switchable attribute “reserve” It is determined that the “system” information does not exist in the label table 22. Based on this result, the node F refers to the path information table 25 (symbol 812), and the upstream node of its own node is “B” from the route “A-B-F-H” of the path ID “1”. The failure information including the path ID “1” is transmitted to the node B based on the routing information stored in itself. As a result, the node B receives the failure information.

  On the other hand, in step S602, if a switchable backup path exists with a path ID corresponding to the IF that detected the failure (S602 → Yes), the failure detection unit 24 of the node sets the IF that detected the failure. By deleting the record of the attribute “working system” of the corresponding path ID, the switching process from the working system path to the protection system path is performed (S604).

  More specifically, referring to FIG. 8, the node B detects the failure by receiving the failure information including the path ID “1” from the node F. Then, the node B refers to the label table 22 (reference numeral 801) and determines that the “standby system” corresponding to the attribute “active system” exists with the path ID “1” included in the failure information. The node B deletes the record with the path ID “1” and the attribute “active” in the label table 22 (reference numeral 801). Subsequently, the node B receives the communication packet with the label “5” received from the upstream node based on the record of the path ID “1” and the attribute “standby” from “5” to “ 4 ”and send from sending IF“ 2 ”. As a result, the communication packet is transmitted to the node C which is the backup path.

  In step S604, the failure detection unit 24, instead of deleting the record of the attribute “working”, for example, information for identifying a record to be referred to when sending the communication packet as a new item in the label table 22 (for example, , A flag such as Yes / No). As a result, the record to be referred to can be identified and switched to the backup path.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to this.

  FIG. 9 is a diagram illustrating a modification of the label table stored in the node. For example, as shown in FIG. 9, the label table 22 may be divided into a label table 22a for storing active information and a label table 22b for storing standby information and stored in the node 1. In this case, the node B, which is a branch node, usually refers to the label table 22a in which information on the active system is stored, and when a failure occurs, the failure detection unit 24 uses the path ID information indicating the failed path. Based on the above, the information in the label table 22b storing the information on the standby system is overwritten on the label table 22a. According to this, since the information of the label table 22a to be normally used can be reduced, the processing load is reduced.

  Further, the failure detection unit 24 may detect not only a failure from a link down or its own IF but also a hardware failure (for example, a failure of the label table 22).

  In the present embodiment, various modifications can be considered for the backup path setting information added to the path setting request. FIG. 10 is a diagram illustrating a modification of the backup path setting information included in the path setting request. Hereinafter, examples (1) to (4) of the backup path setting information illustrated in FIG. 10 will be described.

(1) This is an example of backup path setting information described in the above embodiment.
(2) In this example, a plurality of backup paths can be specified by specifying a plurality of different branch nodes. Thereby, a backup path via different branch nodes can be set.
(3) In this example, a branch node condition is specified instead of directly specifying a branch node, and a plurality of standby end nodes are specified. By specifying a plurality of standby terminal nodes, the branch processing unit 23 selects the shortest path based on information such as cost and delay from a plurality of path candidates when calculating the path of the backup path. Can do.
(4) This is an example in which the above (2) and (3) are combined, and a plurality of branch nodes are designated, and the conditions of the branch nodes are designated.

  As described above, the node 1 can make various designations regarding the backup path when transmitting the path setting request.

  Note that the information on the start point node and the working end node included in the protection path setting information of the path setting request shown in Table 1 can be omitted. In this case, the node 1 acquires information on the start point node and the active system end node from the path setting information in the path setting request. Thereby, the capacity of the path setting request can be reduced and the network load can be reduced.

  In the present embodiment, the attributes stored in the label table 22 have been described as being distinguished between the active system and the standby system. However, for example, the attributes may be distinguished by information indicating priority. When a failure occurs, the priority of the IF record in which the failure has occurred is set to “0”, thereby making it impossible to transmit the communication packet. By distinguishing by priority information, multiple backup paths can be set from one branch node, and even if a failure occurs in the switched target backup path, it is changed to the next priority backup path. Can be switched. In addition, the present invention can be implemented as appropriate without departing from the spirit of the present invention.

  According to the present embodiment, in a network that performs quality assurance, a standby path with different end nodes is set, and the active table is stored in the label table 22 stored in the branch node that branches the active path and the standby path. Information indicating not only the system path but also the backup system path is also stored. When a failure occurs, the relay node refers to the label table 22 and the path information table 25 to notify the failure information to the upstream node of the path where the failure has occurred, and the branch node refers to the label table 22 Switch from the working path where the error occurred to the backup path. This reduces the load on the network and processing and realizes high-speed failure recovery.

  Conventionally, when performing a switching process using a redundant path for quality assurance, there has been a restriction that the end node is shared between the active path and the standby path, but according to the present embodiment, The terminal node can be separated from the backup path, and it is possible to cope with a failure of the terminal node.

  Conventionally, in the case of a network configuration in which the terminal node is divided between the active path and the backup path, if a failure occurs, the branch node performs route calculation by IP routing, so that quick communication recovery is difficult. It was. In addition, communication packet loss until recovery is a problem. However, according to the present embodiment, when a failure occurs, if a branch node refers to the label table 22 and switches, a communication packet is sent to the backup path. As a result, quick communication recovery is possible, and communication packet loss can be minimized.

  Furthermore, in the switching method using multicast, a large processing load is applied to the branch node. However, according to the present embodiment, information on the backup path is stored in the label table 22 of the branch node in advance, so that in the event of a failure, switching processing from the active path to the backup path is performed, so that the backup path is restored. A communication packet is sent out. Therefore, the processing load of the branch node is reduced.

  Further, in the switching method using multicast, the number of backup paths that can be set in one branch node is limited to one. However, according to the present embodiment, the information indicating the priority order in the attribute of the label table 22 of the branch node. Is stored, a plurality of backup paths can be set. As a result, even if a failure occurs in the backup path being used, it is possible to switch to another backup path. In addition, by specifying multiple branch nodes, branch node conditions, or multiple standby path termination points in the backup path setting information included in the path setting request, Flexible switching according to the occurrence location can be realized.

  The node 1 according to the present embodiment can be realized by a program that executes the processing as described above, and can be provided by storing the program in a computer-readable storage medium (CD-ROM or the like). It is. It is also possible to provide the program through a network.

It is a figure which shows the network structural example of the communication system with which the path | route switching system of this embodiment is applied. It is the block diagram which showed the structure of the node which belongs in the core network of FIG. FIG. 2 is a diagram conceptually illustrating a process in which each node transmits a communication packet using a label in the core network illustrated in FIG. 1. It is the flowchart which showed the operation | movement of the node at the time of path setting. It is the sequence diagram which showed the flow of the process of each node in the network shown in FIG. It is a figure which shows the modification of the network shown in FIG. It is the flowchart which showed the operation | movement of the node at the time of a failure. It is a figure which shows notionally the operation | movement of each node at the time of the failure in the network shown in FIG. It is a figure which shows the modification of the label table stored in a node. It is a figure which shows the modification of the protection type | system | group path setting information contained in a path setting request | requirement. It is a figure explaining the conventional network structure.

Explanation of symbols

1 node 2 (2a, 2b) user network 21 path setting unit 22 label table 23 branch processing unit 24 failure detection unit 25 path information table 3 core network A node (path setting node)
B node (branch node)
F node (relay node)
H node (active terminal node)
I node (standby end node)

Claims (6)

In a network, a path switching method for switching from a working path to a backup path having a different termination node with respect to the working path,
A node that becomes a branch point between the active path and the backup path is
A storage unit for storing routing information in the network;
When the routing information of the working path and the protection path is acquired, the interface information for sending the communication packet to the working path and the communication packet are sent to the protection path with reference to the routing information Storing the information on the interface to be associated with each other by using the same path ID in the working path and the backup path in the storage unit;
Detects link down of the link of the own node used by the working path, or receives fault information including the path ID from a downstream node of the working path, thereby causing a fault in the working path via which interface. A step of detecting whether or not
The backup path with the same path ID as the path ID corresponding to the interface where the failure is detected is extracted with reference to the storage unit, and a communication packet is transmitted using the interface of the extracted backup path And a step of performing the path switching. 2. The path switching method according to claim 1, wherein the working path and the backup path use MPLS LSP. In a network composed of a plurality of nodes, in a path switching system including a path composed of a working path and a backup path having a different termination node with respect to the working path,
The node is
A storage unit for storing routing information in the network;
A path ID for identifying a path in the network, a start node of the working path, a termination node of the working path, route information of the working path, and a termination node of a protection path corresponding to the working path , Obtain an active path setting request including at least information of a branch node that branches the active path and the standby path,
Based on the acquired working path setting request, when it is determined that the own node is the starting node of the working path,
Path setting means for transmitting the working path setting request to a downstream node of the working path based on route information of the working path;
When it is determined that the own node is the branch node based on the acquired working path setting request,
The path of the protection path connecting the start path node of the active path to the end node of the protection path included in the active path setting request, the path information of the end node of the active path and the path of the active path The path ID is the same as the path ID of the working path, the start path node of the working path, the end node of the protection path, and the protection path. Branch processing means for creating a backup path setting request including at least the path information of
The working path setting request is transmitted to the downstream node of the working path based on the path information of the working path, and the created protection path setting request is sent based on the path information of the protection path. Path setting means for transmitting to a downstream node of the backup path;
Referring to the routing information, information on an interface for sending a communication packet to the working path and information on an interface for sending a communication packet to the protection path are used as the working path and the protection path. A storage unit that stores the same path ID in association with each other by using the same path ID ;
Detection of a link down of the link of the own node used by the working path, or failure information including the path ID is received from a downstream node of the working path, thereby detecting a failure of the working path. In this case, it is determined with reference to the storage means which interface has failed through the interface, and the same path ID as the path ID corresponding to the determined interface in which the failure has occurred is attached. A failure detection means for sending a communication packet using the backup path interface,
Based on the working path setting request, when the own node is not one of the start node of the working path, the end node of the working path, the end node of the backup path, or the branch node,
Storage means for storing at least route information of the working path included in the working path setting request;
Path setting means for transmitting the working path setting request to a downstream node of the working path based on route information of the working path;
When a failure of the working path is detected, the failure information including the path ID to the upstream node of the working path where the failure has occurred with reference to the path information of the working path stored in the storage unit A failure detection means for transmitting
Based on the working path setting request, when it is determined that the own node is a terminal node of the working path,
Storage means for storing at least route information of the working path included in the working path setting request;
When a failure of the working path is detected, the failure information including the path ID to the upstream node of the working path where the failure has occurred with reference to the path information of the working path stored in the storage unit A path switching system comprising: a failure detection means for transmitting The information included in the acquired working path setting request includes a plurality of end nodes of a protection path for the working path,
The node is
When it is determined that the own node is the branch node based on the acquired working path setting request,
The branch processing means is configured to connect the acquired standby path to a plurality of end nodes so that the branch processing means does not overlap with a path downstream from the own node indicated in the end path of the active path and the path information of the active path. A path is calculated, a termination node and its path information of the path of the backup path that minimizes at least one of cost and delay are selected from the calculated paths of the backup path, and the path of the working path is selected. 4. The backup path setting request including at least the path ID same as the ID, the start node of the working path, the end node of the backup path, and the path information of the backup path is created. Path switching system. In a network composed of a plurality of nodes, a branch node that switches to a standby path corresponding to an active path in the event of a failure,
A storage unit for storing routing information in the network;
Corresponds to a path ID for identifying a path in the network received from another node, a start node of the working path, a terminal node of the working path, route information of the working path, and the working path Based on the active path setting request including at least information on the end node of the protection path, the active path, and the branch node that branches the protection path, the end of the protection path from the start node of the active path The path of the standby path connecting the nodes is calculated so that it does not overlap with the path downstream from the own node indicated in the end node of the working path and the path information of the working path, and the path of the working path same path ID and ID, the start node of the working path, an end node of the backup path, including at least backup path setting request routing information of the backup path A branch processing unit that creates,
The working path setting request is transmitted to the downstream node of the working path based on the path information of the working path, and the created protection path setting request is sent based on the path information of the protection path. A path setting unit for transmitting to a downstream node of the backup path;
Referring to the routing information, information on an interface for sending a communication packet to the working path and information on an interface for sending a communication packet to the protection path are used as the working path and the protection path. The storage unit that stores the same path ID in association with each other by using the same path ID ;
Detection of a link down of the link of the own node used by the working path, or failure information including the path ID is received from a downstream node of the working path, thereby detecting a failure of the working path. In this case, it is determined with reference to the storage unit which interface has failed through the interface, and the same path ID as the path ID corresponding to the determined interface in which the failure has occurred is attached. A branch node comprising: a failure detection unit that transmits a communication packet using an interface of a backup path. The information included in the received working path setting request includes a plurality of end nodes of a protection path for the working path,
The branch processing unit of the branch node is:
A plurality of acquired backup path paths are included so as not to overlap with a downstream path from the own node indicated in the path information of the active path end node and the active path included in the active path setting request. The route to the end node is calculated, and the end node and the route information of the route of the backup path that minimizes at least one of the cost and the delay among the calculated paths of the backup path are selected, and the working A backup path setting request including at least a path ID that is the same as a path ID of a system path, a start node of the active path, a terminal node of the backup path, and route information of the backup path is created. Item 6. The branch node according to item 5 .

Images